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@(?<=\(size )(\d+)\)(?= with)@$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 ($c =~ m/^Reading in filter/) {
1211 return '' unless SHOW_FILTER_READIN_IN;
1215 found_unknown_content($content);
1222 sub handle_loglevel_redirect($) {
1226 if ($c =~ m/^Decoding "([^""]*)"/) {
1228 $req{$t}{'original-destination'} = $1;
1229 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1232 } elsif ($c =~ m/^Checking/) {
1234 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1235 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1237 # TODO: Change colour if really url-decoded
1238 $req{$t}{'decoded-original-destination'} = $1;
1239 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1242 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1244 # pcrs command "s@&from=rss@@" changed \
1245 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1246 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1247 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1248 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1249 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1250 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1251 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1252 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1254 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1256 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1257 # change "http://www.example.org/".
1258 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1259 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1261 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1263 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1264 # XXX: Use URL highlighter
1266 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1268 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1269 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1270 # is already properly formatted.
1271 # XXX: assume the same?
1272 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1274 } elsif ($c =~ m/^Percent-encoding redirect/) {
1276 # Percent-encoding redirect URL: http://www.example.org/\x02
1277 $c = highlight_matched_url($c, '(?<=redirect URL: ).*');
1281 found_unknown_content($c);
1288 sub handle_loglevel_gif_deanimate($) {
1290 my $content = shift;
1292 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1294 my $bytes_from = $1;
1296 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1297 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1298 # XXX: Do we need g in case of ($1 == $2)?
1299 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1301 } elsif ($content =~ m/GIF (not) changed/) {
1303 # Gif-Deanimate: GIF not changed.
1304 return '' if SUPPRESS_GIF_NOT_CHANGED;
1305 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1307 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1309 # failed! (gif parsing)
1310 # XXX: Replace this error message with something less stupid
1311 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1313 } elsif ($content =~ m/^Need to de-chunk first/) {
1315 # Need to de-chunk first
1316 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1318 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1320 # No GIF header found (XXX: Did I ever commit this?)
1321 # failed while parsing 195 134747048 (XXX: never committed)
1323 # Ignore these for now
1327 found_unknown_content($content);
1334 sub handle_loglevel_request($) {
1336 my $content = shift;
1338 if ($content =~ m/crunch! /) {
1340 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1342 # Highlight crunch reasons
1343 foreach my $reason (keys %reason_colours) {
1344 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1346 # Highlight request URL domain and ditch 'crunch!'
1347 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1348 $content =~ s@ crunch!@@;
1350 } elsif ($content =~ m/\[too long, truncated\]$/) {
1352 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1353 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1355 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1357 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1358 $content = h('request_') . $content . h('Standard');
1362 found_unknown_content($content);
1369 sub handle_loglevel_crunch($) {
1371 my $content = shift;
1373 # Highlight crunch reason
1374 foreach my $reason (keys %reason_colours) {
1375 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1378 if ($content =~ m/\[too long, truncated\]$/) {
1380 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1381 # [...]&filter... [too long, truncated]
1382 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1386 # Blocked: http://ads.example.org/
1387 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1393 sub handle_loglevel_connect($) {
1397 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1399 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1401 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1402 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1404 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1406 # connect to: www.example.org.noconnect failed: Operation not permitted
1408 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1410 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1412 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1414 # Connect: to www.nzherald.co.nz successful
1415 # Connect: to archiv.radiotux.de
1417 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1418 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1420 } elsif ($c =~ m/^to ([^\s]*)$/) {
1422 # Connect: to lists.sourceforge.net:443
1424 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1426 } elsif ($c =~ m/^[Aa]ccepted connection from .*/ or
1430 # Accepted connection from 10.0.0.1 on socket 5
1431 # Privoxy between 3.0.20 and 3.0.6:
1432 # accepted connection from 10.0.0.1( on socket 5)?
1433 # Privoxy 3.0.6 and earlier just say:
1435 $c = highlight_matched_host($c, '(?<=connection from )[^ ]*');
1436 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1438 } elsif ($c =~ m/^Closing client socket/) {
1440 # Closing client socket 5. Keep-alive: 0, Socket alive: 1. Data available: 0.
1441 # Privoxy 3.0.20 and later
1442 # Closing client socket 8. Keep-alive: 1. Socket alive: 0. Data available: 0. \
1443 # Configuration file change detected: 0. Requests received: 11.
1445 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1446 $c = highlight_matched_pattern($c, 'Number', '(?<=Keep-alive: )\d+');
1447 $c = highlight_matched_pattern($c, 'Number', '(?<=Socket alive: )\d+');
1448 $c = highlight_matched_pattern($c, 'Number', '(?<=available: )\d+');
1449 $c = highlight_matched_pattern($c, 'Number', '(?<=detected: )\d+');
1450 $c = highlight_matched_pattern($c, 'Number', '(?<=received: )\d+');
1452 } elsif ($c =~ m/^write header to: .* failed:/) {
1454 # write header to: 10.0.0.1 failed: Broken pipe
1456 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1457 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1459 } elsif ($c =~ m/^write header to client failed:/) {
1461 # write header to client failed: Broken pipe
1463 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1465 } elsif ($c =~ m/^socks4_connect:/) {
1467 # socks4_connect: SOCKS request rejected or failed.
1468 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1470 } elsif ($c =~ m/^Listening for new connections/ or
1471 $c =~ m/^accept connection/) {
1473 # Privoxy versions above 3.0.6 say:
1474 # Listening for new connections ...
1475 # earlier versions say:
1476 # accept connection ...
1479 } elsif ($c =~ m/^accept failed:/) {
1481 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1483 } elsif ($c =~ m/^Overriding forwarding settings/) {
1485 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1486 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1488 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1490 # Denying suspicious CONNECT request from 10.0.0.1
1491 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1493 } elsif ($c =~ m/^socks5_connect:/) {
1495 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1497 } elsif ($c =~ m/^Created new connection to/) {
1499 # Created new connection to www.privoxy.org:80 on socket 11.
1500 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1501 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1503 } elsif ($c =~ m/^Found reusable socket/) {
1505 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1507 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1508 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1509 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1510 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1511 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1512 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1513 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1514 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 } elsif ($c =~ m/^Marking open socket/) {
1518 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1519 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1520 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1521 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1523 } elsif ($c =~ m/^No reusable/) {
1525 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1526 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1528 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1530 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1531 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1533 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1534 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1535 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1537 } elsif ($c =~ m/^Socket/) {
1539 # Socket 16 already forgotten or never remembered.
1540 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1542 } elsif ($c =~ m/^The connection to/) {
1544 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1546 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1547 # Timeout is: 1. Assumed latency: 4.
1548 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1549 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1550 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1551 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1553 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1555 } elsif ($c =~ m/^Stopped waiting for the request line/ or
1556 $c =~ m/^No request line on socket \d received in time/ or
1557 $c =~ m/^The client side of the connection on socket \d/) {
1559 # Stopped waiting for the request line. Timeout: 121.
1560 # Privoxy 3.0.19 and later:
1561 # No request line on socket 5 received in time. Timeout: 1.
1562 # The client side of the connection on socket 5 got closed \
1563 # without sending a complete request line.
1564 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1565 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 } elsif ($c =~ m/^Waiting for \d/) {
1569 # Waiting for 1 connections to timeout.
1570 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1572 } elsif ($c =~ m/^Initialized/) {
1574 # Initialized 20 socket slots.
1575 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1577 } elsif ($c =~ m/^Done reading from server/) {
1579 # Done reading from server. Expected content length: 24892. \
1580 # Actual content length: 24892. Most recently received: 4412.
1582 # Done reading from server. Expected content length: 24892. \
1583 # Actual content length: 24892. Bytes most recently read: 4412.
1584 # Done reading from server. Content length: 6018 as expected. \
1585 # Bytes most recently read: 294.
1586 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1587 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1588 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1590 } elsif ($c =~ m/^Continuing buffering (?:server )?headers/) {
1592 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1593 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1594 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1595 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1596 # 3.0.15 up to 3.0.19:
1597 # Continuing buffering headers. Bytes most recently read: 498.
1598 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1600 # Continuing buffering server headers from socket 5. Bytes most recently read: 498.
1601 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1603 } elsif ($c =~ m/^Received \d+ bytes while/) {
1605 # Received 206 bytes while expecting 12103.
1606 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1609 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1611 # Connection from 81.163.28.218 dropped due to ACL
1612 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1613 # Connection from 192.168.2.1 on 127.0.1.1:8118 (socket 3) dropped due to ACL
1614 $c = highlight_matched_host($c, '(?<=onnection from )[\d.:]+');
1615 $c = highlight_matched_host($c, '(?<=on )[\d.:]+');
1616 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1618 } elsif ($c =~ m/^(?:Reusing|Closing) server socket / or
1619 $c =~ m/^No additional client request/) {
1621 # Reusing server socket 4. Opened for 10.0.0.1.
1622 # Closing server socket 2. Opened for 10.0.0.1.
1623 # No additional client request received in time. \
1624 # Closing server socket 4, initially opened for 10.0.0.1.
1625 # No additional client request received in time on socket 29.
1626 # Privoxy 3.0.20 and later
1627 # Reusing server socket 7 connected to www.privoxy.org. Total requests: 2.
1628 # Closing server socket 6 connected to d.asset.soup.io. Keep-alive: 0.\
1629 # Tainted: 1. Socket alive: 1. Timeout: 60. Configuration file change detected: 0.
1631 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1632 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.)');
1633 $c = highlight_matched_host($c, '(?<=connected to )[^\s]+(?=\.)');
1634 for my $number_pattern ('requests', 'Keep-alive', 'Tainted', ' alive', 'Timeout', 'detected') {
1635 $c = highlight_matched_pattern($c, 'Number', '(?<='. $number_pattern . ': )\d+');
1638 } elsif ($c =~ m/^Connected to /) {
1640 # Connected to tor-jail[10.0.0.2]:9050.
1642 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1643 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1644 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1646 } elsif ($c =~ m/^Could not connect to /) {
1648 # Could not connect to [10.0.0.1]:80.
1650 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1651 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1653 } elsif ($c =~ m/^Waiting for the next client request/ or
1654 $c =~ m/^The connection on server socket/ or
1655 $c =~ m/^Client request (?:\d+ )?(?:arrived in time|has been pipelined) /) {
1657 # Waiting for the next client request on socket 3. Keeping the server \
1658 # socket 12 to a.fsdn.com open.
1659 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1660 # Privoxy 3.0.20 and later:
1661 # Client request 4 arrived in time on socket 7.
1662 # Used by Privoxy 3.0.18 and 3.0.19:
1663 # Client request arrived in time on socket 21.
1664 # Used by earlier version:
1665 # Client request arrived in time or the client closed the connection on socket 12.
1666 # Client request 8 has been pipelined on socket 7 and the socket is still alive.
1668 $c =~ s@(?<=request )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1669 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1670 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1671 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1673 } elsif ($c =~ m/^Marking the server socket/) {
1675 # Marking the server socket 7 tainted.
1677 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1679 } elsif ($c =~ m/^Reduced expected bytes to /) {
1681 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1682 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1683 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1685 } elsif ($c =~ m/^The client closed socket /) {
1687 # The client closed socket 2 while the server socket 4 is still open.
1688 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1689 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1691 } elsif ($c =~ m/^Expected client content length set /) {
1693 # Expected client content length set to 667325411 after reading 4999 bytes.
1694 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1695 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1697 } elsif ($c =~ m/^Reducing expected bytes to /) {
1699 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1700 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1701 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1703 } elsif ($c =~ m/^Waiting for up to /) {
1705 # Waiting for up to 4999 bytes from the client.
1706 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1708 } elsif ($c =~ m/^Optimistically sending /) {
1710 # Optimistically sending 318 bytes of client headers intended for www.privoxy.org
1711 $c =~ s@(?<=sending )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1712 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1714 } elsif ($c =~ m/^Stopping to watch the client socket/) {
1716 # Stopping to watch the client socket. There's already another request waiting.
1717 # Privoxy 3.0.20 and later:
1718 # Stopping to watch the client socket 5. There's already another request waiting.
1719 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1721 } elsif ($c =~ m/^Drained \d+ bytes before closing/) {
1723 # Drained 180 bytes before closing socket 6
1724 $c =~ s@(?<=Drained )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1725 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1727 } elsif ($c =~ m/^Tainting client socket/ or
1728 $c =~ m/^Failed to shutdown socket/) {
1730 # Tainting client socket 7 due to unread data.
1731 # Failed to shutdown socket 11: Connection reset by peer
1733 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1735 } elsif ($c =~ m/^Shifting \d+ pipelined bytes/) {
1737 # Shifting 360 pipelined bytes by 360 bytes
1738 $c =~ s@(?<=Shifting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1739 $c =~ s@(?<=by )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1741 } elsif ($c =~ m/^Flushed (\d+) bytes of request body while expecting (\d+)/) {
1743 # Flushed 30 bytes of request body while expecting 30
1744 $c =~ s@(?<=Flushed )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1745 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1747 } elsif ($c =~ m/^Performing the TLS\/SSL handshake with client. Hash of host:/) {
1749 # Performing the TLS/SSL handshake with client. Hash of host: bab5296b25e256c7b06b92b17b56bcae
1750 $c = highlight_matched_host($c, '(?<=Hash of host: ).+');
1752 } elsif ($c =~ m/^Forwarding \d+ bytes of encrypted POST data/) {
1754 # Forwarding 1954 bytes of encrypted POST data
1755 $c =~ s@(?<=Forwarding )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1757 } elsif ($c =~ m/^Forwarded the last \d+ bytes/) {
1759 # Forwarded the last 1954 bytes
1760 $c =~ s@(?<=the last )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1762 } elsif ($c =~ m/^Looks like we / or
1763 $c =~ m/^Unsetting keep-alive flag/ or
1764 $c =~ m/^No connections to wait/ or
1765 $c =~ m/^Complete client request received/ or
1766 $c =~ m/^Possible pipeline attempt detected./ or
1767 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1768 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1769 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1770 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1771 $c =~ m/^There better be a request body./ or
1772 $c =~ m/^Done reading from the client\.$/) {
1774 # Looks like we reached the end of the last chunk. We better stop reading.
1775 # Looks like we read the end of the last chunk together with the server \
1776 # headers. We better stop reading.
1777 # Looks like we got the last chunk together with the server headers. \
1778 # We better stop reading.
1779 # Unsetting keep-alive flag.
1780 # No connections to wait for left.
1781 # Client request arrived in time or the client closed the connection.
1782 # Complete client request received
1783 # Possible pipeline attempt detected. The connection will not be \
1784 # kept alive and we will only serve the first request.
1785 # POST request detected. The connection will not be kept alive.
1786 # The server still wants to talk, but the client hung up on us.
1787 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1788 # There might be a request body. The connection will not be kept alive.
1789 # Privoxy 3.0.20 and later
1790 # There better be a request body.
1791 # Done reading from the client.
1795 found_unknown_content($c);
1803 sub handle_loglevel_info($) {
1807 if ($c =~ m/^Rewrite detected:/) {
1809 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1810 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1812 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1813 $c =~ m/^Compressed content detected/ or
1814 $c =~ m/^SDCH-compressed content detected/ or
1817 # Decompressing deflated iob: 117
1818 # Decompression didn't result in any content.
1819 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1820 # with zlib support or enable the prevent-compression action.
1821 # SDCH-compressed content detected, content filtering disabled.\
1822 # Consider suppressing SDCH offers made by the client.
1823 # Tagger 'complete-url' created empty tag. Ignored.
1827 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1829 # loading configuration file '/usr/local/etc/privoxy/config':
1830 # Reloading configuration file '/usr/local/etc/privoxy/config'
1831 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1833 } elsif ($c =~ m/^Loading (actions|filter|trust) file: /) {
1835 # Loading actions file: /usr/local/etc/privoxy/default.action
1836 # Loading filter file: /usr/local/etc/privoxy/default.filter
1837 # Loading trust file: /usr/local/etc/privoxy/trust
1839 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1841 } elsif ($c =~ m/^exiting by signal/) {
1843 # exiting by signal 15 .. bye
1844 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1846 } elsif ($c =~ m/^Privoxy version/) {
1848 # Privoxy version 3.0.7
1849 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1851 } elsif ($c =~ m/^Program name: /) {
1853 # Program name: /usr/local/sbin/privoxy
1854 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1856 } elsif ($c =~ m/^Listening on port /) {
1858 # Listening on port 8118 on IP address 10.0.0.1
1859 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1860 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1862 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1864 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1865 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1867 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1869 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1870 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1872 # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
1873 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1875 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1876 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1877 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1878 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1879 $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
1881 } elsif ($c =~ m/^Status code/) {
1883 # Status code 304 implies no body.
1884 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1886 } elsif ($c =~ m/^Method/) {
1888 # Method HEAD implies no body.
1889 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1891 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1893 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1894 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1895 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1897 } elsif ($c =~ m/^File modification detected: /) {
1899 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1900 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1902 } elsif ($c =~ m/^No logfile configured/ or
1903 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1904 $c =~ m/^Invalid \"chunked\" transfer/ or
1905 $c =~ m/^Support for/ or
1906 $c =~ m/^Flushing header and buffers/ or
1907 $c =~ m/^Can not resolve/
1910 # No logfile configured. Please enable it before reporting any problems.
1911 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1912 # response or even no response at all.
1913 # No logfile configured. Logging disabled.
1914 # Invalid "chunked" transfer encoding detected and ignored.
1915 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1916 # known not to work properly in some situations.
1917 # Flushing header and buffers. Stepping back from filtering.
1918 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1922 found_unknown_content($c);
1929 sub handle_loglevel_cgi($) {
1933 if ($c =~ m/^Granting access to/) {
1935 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1937 } elsif ($c =~ m/^Substituting: s(.)/) {
1939 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1940 # XXX: prone to span several lines
1943 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1944 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1945 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1951 sub handle_loglevel_force($) {
1955 if ($c =~ m/^Ignored force prefix in request:/) {
1957 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1958 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1959 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1961 } elsif ($c =~ m/^Enforcing request:/) {
1963 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1964 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1968 found_unknown_content($c);
1975 sub handle_loglevel_error($) {
1979 if ($c =~ m/^(?:Empty|No) server or forwarder response received on socket \d+\./) {
1981 # Empty server or forwarder response received on socket 4.
1982 # Empty server or forwarder response received on socket 3. \
1983 # Closing client socket 15 without sending data.
1984 # Used by Privoxy 3.0.18 and later:
1985 # No server or forwarder response received on socket 8. \
1986 # Closing client socket 10 without sending data.
1988 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1989 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1991 } elsif ($c =~ m/^Didn't receive data in time:/) {
1993 # Didn't receive data in time: a.fsdn.com:443
1994 $c =~ s@(?<=in time: )(.*)@$h{'destination'}$1$h{'Standard'}@;
1997 # XXX: There are probably more messages that deserve highlighting.
2003 sub handle_loglevel_ignore($) {
2007 sub gather_loglevel_clf_stats($) {
2009 my $content = shift;
2010 my ($method, $resource, $http_version, $status_code, $size);
2014 # +0200] "GET https://www.youtube.com/watch?v=JmcA9LIIXWw HTTP/1.1" 200 68004
2015 $content =~ m/^[+-]\d{4}\] "(\w+) (.+) (HTTP\/\d\.\d)" (\d+) (\d+)/;
2022 unless (defined $method) {
2023 print("Failed to parse: $content\n");
2026 $stats{'method'}{$method}++;
2027 if ($cli_options{'url-statistics-threshold'} != 0) {
2028 $stats{'resource'}{$resource}++;
2030 $stats{'http-version'}{$http_version}++;
2032 if ($cli_options{'host-statistics-threshold'} != 0) {
2033 $resource =~ m@(?:http[s]://)([^/]+)/?@;
2034 $stats{'hosts'}{$1}++;
2036 $stats{'content-size-total'} += $size;
2037 $stats{'status-code'}{$status_code}++;
2038 $stats{requests_clf}++;
2041 sub gather_loglevel_request_stats($$) {
2049 sub gather_loglevel_crunch_stats($$) {
2056 if ($c =~ m/^Redirected:/) {
2057 # Redirected: http://www.example.org/http://p.p/
2058 $stats{'fast-redirections'}++;
2060 } elsif ($c =~ m/^Blocked:/) {
2061 # Blocked: blogger.googleusercontent.com:443
2062 $stats{'blocked'}++;
2064 } elsif ($c =~ m/^Connection timeout:/) {
2065 # Connection timeout: http://c.tile.openstreetmap.org/18/136116/87842.png
2066 $stats{'connection-timeout'}++;
2068 } elsif ($c =~ m/^Connection failure:/) {
2069 # Connection failure: http://127.0.0.1:8080/
2070 $stats{'connection-failure'}++;
2075 sub gather_loglevel_error_stats($$) {
2082 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
2084 # Empty server or forwarder response received on socket 4.
2085 $stats{'empty-responses'}++;
2086 if ($thread_data{$thread}{'new_connection'}) {
2087 $stats{'empty-responses-on-new-connections'}++;
2089 $stats{'empty-responses-on-reused-connections'}++;
2094 sub gather_loglevel_connect_stats($$) {
2096 my ($c, $thread) = @_;
2100 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
2102 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
2103 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
2105 } elsif ($c =~ m/^to ([^\s]*)$/) {
2107 # Connect: to lists.sourceforge.net:443
2109 $thread_data{$thread}{'forwarder'} = 'direct connection';
2111 } elsif ($c =~ m/^Created new connection to/) {
2113 # Created new connection to www.privoxy.org:80 on socket 11.
2115 $thread_data{$thread}{'new_connection'} = 1;
2117 } elsif ($c =~ m/^Reusing server socket \d./ or
2118 $c =~ m/^Found reusable socket/) {
2120 # Reusing server socket 4. Opened for 10.0.0.1.
2121 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
2123 $thread_data{$thread}{'new_connection'} = 0;
2124 $stats{'reused-connections'}++;
2126 } elsif ($c =~ m/^Closing client socket \d+. .* Requests received: (\d+)\.$/) {
2128 # Closing client socket 12. Keep-alive: 1. Socket alive: 1. Data available: 0. \
2129 # Configuration file change detected: 0. Requests received: 14.
2131 $stats{'client-requests-on-connection'}{$1}++;
2132 $stats{'closed-client-connections'}++;
2136 sub gather_loglevel_header_stats($$) {
2138 my ($c, $thread) = @_;
2142 if ($c =~ m/^A HTTP\/1\.1 response without/ or
2143 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
2145 # A HTTP/1.1 response without Connection header implies keep-alive.
2146 # Keeping the server header 'Connection: keep-alive' around.
2147 $stats{'server-keep-alive'}++;
2156 'server-keep-alive' => 0,
2157 'reused-connections' => 0,
2158 'empty-responses' => 0,
2159 'empty-responses-on-new-connections' => 0,
2160 'empty-responses-on-reused-connections' => 0,
2161 'fast-redirections' => 0,
2163 'connection-failure' => 0,
2164 'connection-timeout' => 0,
2165 'reused-connections' => 0,
2166 'server-keep-alive' => 0,
2167 'closed-client-connections' => 0,
2168 'content-size-total' => 0,
2170 $stats{'client-requests-on-connection'}{1} = 0;
2173 sub get_percentage($$) {
2177 # If small is 0 the percentage is always 0%.
2178 # Make sure it works even if big is 0 as well.
2179 return "0.00%" if ($small eq 0);
2181 # Prevent division by zero.
2182 # XXX: Is this still supposed to be reachable?
2183 return "NaN" if ($big eq 0);
2185 return sprintf("%.2f%%", $small / $big * 100);
2192 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2193 my $client_requests_checksum = 0;
2196 if ($stats{requests_clf} && $stats{requests}
2197 && $stats{requests_clf} != $stats{requests}) {
2198 print "Inconsistent request counts: " . $stats{requests} . "/" . $stats{requests_clf} . "\n";
2201 # To get the total number of requests we can use either the number
2202 # of Common-Log-Format lines or the number of "Request:" messages.
2203 # We prefer the number of CLF lines if available because using
2204 # it works when analysing old log files from Privoxy versions before 3.0.29.
2205 # In Privoxy 3.0.28 and earlier "Request:" messages excluded
2206 # crunched messages.
2207 $requests_total = $stats{requests_clf} ? $stats{requests_clf} : $stats{requests};
2209 if ($requests_total eq 0) {
2210 print "No requests yet.\n";
2214 print "Client requests total: " . $requests_total . "\n";
2215 if ($stats{crunches}) {
2216 my $outgoing_requests = $requests_total - $stats{crunches};
2217 print "Crunches: " . $stats{crunches} . " (" .
2218 get_percentage($requests_total, $stats{crunches}) . ")\n";
2219 print "Blocks: " . $stats{'blocked'} . " (" .
2220 get_percentage($requests_total, $stats{'blocked'}) . ")\n";
2221 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2222 get_percentage($requests_total, $stats{'fast-redirections'}) . ")\n";
2223 print "Connection timeouts: " . $stats{'connection-timeout'} . " (" .
2224 get_percentage($requests_total, $stats{'connection-timeout'}) . ")\n";
2225 print "Connection failures: " . $stats{'connection-failure'} . " (" .
2226 get_percentage($requests_total, $stats{'connection-failure'}) . ")\n";
2227 print "Outgoing requests: " . $outgoing_requests . " (" .
2228 get_percentage($requests_total, $outgoing_requests) . ")\n";
2230 print "No crunches detected. Is 'debug 1024' enabled?\n";
2233 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2234 get_percentage($requests_total, $stats{'server-keep-alive'}) . ")\n";
2235 print "New outgoing connections: " . $new_connections . " (" .
2236 get_percentage($requests_total, $new_connections) . ")\n";
2237 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2238 get_percentage($requests_total, $stats{'reused-connections'}) .
2239 "; server offers accepted: " .
2240 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2241 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2242 get_percentage($requests_total, $stats{'empty-responses'}) . ")\n";
2243 print "Empty responses on new connections: "
2244 . $stats{'empty-responses-on-new-connections'} . " (" .
2245 get_percentage($requests_total, $stats{'empty-responses-on-new-connections'})
2247 print "Empty responses on reused connections: " .
2248 $stats{'empty-responses-on-reused-connections'} . " (" .
2249 get_percentage($requests_total, $stats{'empty-responses-on-reused-connections'}) .
2251 print "Client connections: " . $stats{'closed-client-connections'} . "\n";
2252 if ($stats{'content-size-total'}) {
2253 print "Bytes of content transfered to the client: " . $stats{'content-size-total'} . "\n";
2255 my $lines_printed = 0;
2256 print "Client requests per connection distribution:\n";
2257 foreach my $client_requests (sort {
2258 $stats{'client-requests-on-connection'}{$b} <=> $stats{'client-requests-on-connection'}{$a}}
2259 keys %{$stats{'client-requests-on-connection'}
2262 my $count = $stats{'client-requests-on-connection'}{$client_requests};
2263 $client_requests_checksum += $count * $client_requests;
2264 if ($cli_options{'show-complete-request-distribution'} or ($lines_printed < 10)) {
2265 printf "%8d: %d\n", $count, $client_requests;
2269 unless ($cli_options{'show-complete-request-distribution'}) {
2270 printf "Enable --show-complete-request-distribution to get less common numbers as well.\n";
2272 # Due to log rotation we may not have a complete picture for all the requests
2273 printf "Improperly accounted requests: ~%d\n", abs($requests_total - $client_requests_checksum);
2275 if (exists $stats{method}) {
2276 print "Method distribution:\n";
2277 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2278 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2281 print "Method distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2283 if (exists $stats{'http-version'}) {
2284 print "Client HTTP versions:\n";
2285 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2286 printf "%8d : %-8s\n", $stats{'http-version'}{$http_version}, $http_version;
2289 print "HTTP version distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2291 if (exists $stats{'status-code'}) {
2292 print "HTTP status codes:\n";
2293 foreach my $status_code (sort {$stats{'status-code'}{$b} <=> $stats{'status-code'}{$a}} keys %{$stats{'status-code'}}) {
2294 printf "%8d : %-8d\n", $stats{'status-code'}{$status_code}, $status_code;
2297 print "Status code distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2300 if ($cli_options{'url-statistics-threshold'} == 0) {
2301 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2303 print "Requested URLs:\n";
2304 foreach my $resource (sort {$stats{'resource'}{$b} <=> $stats{'resource'}{$a}} keys %{$stats{'resource'}}) {
2305 if ($stats{'resource'}{$resource} < $cli_options{'url-statistics-threshold'}) {
2306 print "Skipped statistics for URLs below the treshold.\n";
2309 printf "%d : %s\n", $stats{'resource'}{$resource}, $resource;
2313 if ($cli_options{'host-statistics-threshold'} == 0) {
2314 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2316 print "Requested Hosts:\n";
2317 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2318 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2319 print "Skipped statistics for Hosts below the treshold.\n";
2322 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2328 ################################################################################
2329 # Functions that actually print stuff
2330 ################################################################################
2332 sub print_clf_message() {
2334 our ($ip, $timestamp, $request_line, $status_code, $size);
2337 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2339 # Rebuild highlighted
2340 $output .= $h{'Number'} . $ip . $h{'Standard'};
2342 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2344 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2346 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2348 $output .= $h{'Number'} . $size . $h{'Standard'};
2349 $output .= $line_end;
2354 sub print_non_clf_message($) {
2356 my $content = shift;
2357 my $msec_string = $no_msecs_mode ? '' : '.' . $req{$t}{'msecs'};
2358 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2360 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2363 . $time_colours[$time_colour_index % 2]
2364 . $req{$t}{'time-stamp'}
2366 . $h{Standard} . " "
2367 . $thread_colours{$t}
2371 . $h{$req{$t}{'log-level'}}
2372 . $req{$t}{'log-level'}
2379 sub shorten_thread_id($) {
2381 my $thread_id = shift;
2383 our %short_thread_ids;
2386 unless (defined $short_thread_ids{$thread_id}) {
2387 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2390 return $short_thread_ids{$thread_id}
2395 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2397 my $last_thread = 0;
2398 my $last_timestamp = 0;
2399 my $filters_that_did_nothing;
2402 $time_colour = paint_it('white');
2404 my %log_level_handlers = (
2405 'Re-Filter' => \&handle_loglevel_re_filter,
2406 'Header' => \&handle_loglevel_header,
2407 'Connect' => \&handle_loglevel_connect,
2408 'Redirect' => \&handle_loglevel_redirect,
2409 'Request' => \&handle_loglevel_request,
2410 'Crunch' => \&handle_loglevel_crunch,
2411 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2412 'Info' => \&handle_loglevel_info,
2413 'CGI' => \&handle_loglevel_cgi,
2414 'Force' => \&handle_loglevel_force,
2415 'Error' => \&handle_loglevel_error,
2416 'Fatal error' => \&handle_loglevel_ignore,
2417 'Writing' => \&handle_loglevel_ignore,
2418 'Received' => \&handle_loglevel_ignore,
2419 'Actions' => \&handle_loglevel_ignore,
2420 'Unknown log level' => \&handle_loglevel_ignore,
2425 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2426 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2427 $req{$t}{'day'} = $day = $1;
2428 $req{$t}{'time-stamp'} = $time_stamp = $2;
2429 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2430 $req{$t}{'log-level'} = $log_level = $5;
2431 $req{$t}{'content'} = $content = $c = $6;
2432 $req{$t}{'log-message'} = $_;
2433 $no_special_header_highlighting = 0;
2435 if (defined($log_level_handlers{$log_level})) {
2437 $content = $log_level_handlers{$log_level}($content);
2441 die "No handler found for log level \"$log_level\"\n";
2444 # Highlight Truncations
2445 if (length($_) > 4000) {
2446 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2449 next unless $content;
2451 # Register threads to keep the colour constant
2452 if (!defined($thread_colours{$thread})) {
2453 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2454 $thread_colour_index++;
2457 # Switch timestamp colour if timestamps differ
2458 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2459 debug_message("Tick tack!") if DEBUG_TICKS;
2460 $time_colour = $time_colours[$time_colour_index % 2];
2461 $time_colour_index++;
2462 $last_msecs = $msecs;
2463 $last_timestamp = $time_stamp;
2466 $last_thread = $thread;
2468 print_non_clf_message($content);
2470 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2472 # LOG_LEVEL_CLF lines look like this
2473 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2474 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2475 # HTTP/1.1" 403 1730
2476 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2478 print_clf_message();
2482 # Some Privoxy log messages span more than one line,
2483 # usually to dump lots of content that doesn't need any syntax highlighting.
2484 # XXX: add mechanism to forward these lines to the right handler anyway.
2486 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2487 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2495 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2496 my $strict_checks = cli_option_is_set('strict-checks');
2497 my %log_level_handlers = (
2498 'Connect:' => \&gather_loglevel_connect_stats,
2499 'Crunch:' => \&gather_loglevel_crunch_stats,
2500 'Error:' => \&gather_loglevel_error_stats,
2501 'Header:' => \&gather_loglevel_header_stats,
2502 'Request:' => \&gather_loglevel_request_stats,
2504 my %ignored_log_levels = (
2505 'Actions:' => \&handle_loglevel_ignore,
2506 'CGI:' => \&handle_loglevel_ignore,
2507 'Fatal error:' => \&handle_loglevel_ignore,
2508 'Force:' => \&handle_loglevel_ignore,
2509 'Gif-Deanimate:' => \&handle_loglevel_ignore,
2510 'Info:' => \&handle_loglevel_ignore,
2511 'Re-Filter:' => \&handle_loglevel_ignore,
2512 'Received:' => \&handle_loglevel_ignore,
2513 'Redirect:' => \&handle_loglevel_ignore,
2514 'Unknown log level:' => \&handle_loglevel_ignore,
2515 'Writing:' => \&handle_loglevel_ignore,
2519 (undef, $time_stamp, $thread, $log_level, $content) = split(/ /, $_, 5);
2522 next if (not defined($log_level));
2524 if ($time_stamp eq "-") {
2526 gather_loglevel_clf_stats($content);
2528 } elsif (defined($log_level_handlers{$log_level})) {
2530 $content = $log_level_handlers{$log_level}($content, $thread);
2532 } elsif ($strict_checks and not defined($ignored_log_levels{$log_level})) {
2534 die "No handler found for: $_";
2542 sub unbreak_lines_only_loop() {
2543 my $log_messages_reached = 0;
2547 # Log level other than LOG_LEVEL_CLF?
2548 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2550 m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2551 $log_messages_reached = 1;
2556 $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
2561 print "\n" unless $log_messages_reached;
2566 sub VersionMessage {
2567 my $version_message;
2569 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2570 $version_message .= 'https://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2572 print $version_message;
2575 sub get_cli_options() {
2577 our %cli_options = (
2578 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2579 'title' => CLI_OPTION_TITLE,
2580 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2581 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2582 'no-msecs' => CLI_OPTION_NO_MSECS,
2583 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2584 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2585 'statistics' => CLI_OPTION_STATISTICS,
2586 'strict-checks' => CLI_OPTION_STRICT_CHECKS,
2587 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2588 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2589 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2590 'show-complete-request-distribution' => CLI_OPTION_SHOW_COMPLETE_REQUEST_DISTRIBUTION,
2594 'html-output' => \$cli_options{'html-output'},
2595 'title' => \$cli_options{'title'},
2596 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2597 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2598 'no-msecs' => \$cli_options{'no-msecs'},
2599 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2600 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2601 'statistics' => \$cli_options{'statistics'},
2602 'strict-checks' => \$cli_options{'strict-checks'},
2603 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2604 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
2605 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
2606 'show-complete-request-distribution' => \$cli_options{'show-complete-request-distribution'},
2607 'version' => sub { VersionMessage && exit(0) },
2611 $html_output_mode = cli_option_is_set('html-output');
2612 $no_msecs_mode = cli_option_is_set('no-msecs');
2613 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2614 $line_end = get_line_end();
2625 Options and their default values if they have any:
2626 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2630 [--no-syntax-highlighting]
2631 [--shorten-thread-ids]
2632 [--show-ineffective-filters]
2633 [--show-complete-request-distribution]
2635 [--unbreak-lines-only]
2636 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2637 [--title $cli_options{'title'}]
2639 see "perldoc $0" for more information
2645 ################################################################################
2647 ################################################################################
2651 set_background(DEFAULT_BACKGROUND);
2652 prepare_our_stuff();
2656 # XXX: should explicitly reject incompatible argument combinations
2657 if (cli_option_is_set('unbreak-lines-only')) {
2658 unbreak_lines_only_loop();
2659 } elsif (cli_option_is_set('statistics')) {
2672 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2676 B<privoxy-log-parser> [B<--html-output>]
2677 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2678 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2679 [B<--url-statistics-threshold>] [B<--version>]
2683 B<privoxy-log-parser> reads Privoxy log messages and
2685 - syntax-highlights recognized lines,
2687 - reformats some of them for easier comprehension,
2689 - filters out less useful messages, and
2691 - (in some cases) calculates additional information,
2692 like the compression ratio or how a filter affected
2695 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2696 without getting confused by the resulting amount of output. For example for
2697 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2698 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2699 will hide the "filter foo caused 0 hits" message.
2703 [B<--host-statistics-threshold>] Only show the request count for a host
2704 if it's above or equal to the given threshold. If the threshold is 0, host
2705 statistics are disabled.
2707 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2708 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2709 This option is only intended to make embedding log excerpts in web pages easier.
2710 It does not escape any input!
2712 [B<--no-msecs>] Don't expect milisecond resolution
2714 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2715 the filtered output is piped into less in which case the ANSI control
2716 codes don't work, or if the terminal itself doesn't support the control
2719 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2720 Note that the mapping from thread ids to shortened ids is created at run-time
2721 and thus varies with the input.
2723 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2724 that didn't modify the content.
2726 [B<--show-complete-request-distribution>] Show the complete client request
2727 distribution in the B<--statistics> output. Without this option only the
2728 ten most common numbers are shown.
2730 [B<--statistics>] Gather various statistics instead of syntax highlighting
2731 log messages. This is an experimental feature, if the results look wrong
2732 they very well might be. Also note that the results are pretty much guaranteed
2733 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2735 [B<--strict-checks>] When generating statistics, look more careful at the
2736 input data and abort if it is unexpected, even if it doesn't affect the
2737 results. Significantly slows the parsing down and is not expected to catch
2738 any problems that matter.
2739 When highlighting, print warnings in case of unknown messages which can't be
2740 properly highlighted.
2742 [B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
2743 interestingly configured mail client and thus are no longer recognized properly.
2744 Only fixes some breakage, but may be good enough or at least better than nothing.
2745 Doesn't do anything else, so you probably want to pipe the output into
2746 B<privoxy-log-parser> again.
2748 [B<--url-statistics-threshold>] Only show the request count for a resource
2749 if it's above or equal to the given threshold. If the threshold is 0, URL
2750 statistics are disabled.
2752 [B<--version>] Print version and exit.
2756 To monitor a log file:
2758 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2760 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2761 or if the log won't get rotated anyway. The log file location depends
2762 on your system (Doh!).
2764 To monitor Privoxy without having it write to a log file:
2766 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2768 Again, the config file location depends on your system. Output redirection
2769 depends on your shell, the above works with bourne shells.
2771 To read a processed Privoxy log file from top to bottom, letting the content
2772 scroll by slightly faster than you can read:
2774 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2776 This is probably only useful to fill screens in the background of haxor movies.
2780 Syntax highlighting with ANSI escape sequences will look strange
2781 if your background color isn't black.
2783 Some messages aren't recognized yet and will not be fully highlighted.
2785 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2786 using earlier Privoxy versions will probably result in an increased amount
2787 of unrecognized log lines.
2789 Privoxy's log files tend to be rather large. If you use HTML
2790 highlighting some browsers can't handle them, get confused and
2791 will eventually crash because of segmentation faults or unexpected
2792 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2797 Many settings can't be controlled through command line options yet.
2805 Fabian Keil <fk@fabiankeil.de>