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 => '0.9.2',
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_KEEP_DATE => 0,
56 CLI_OPTION_NO_EMBEDDED_CSS => 0,
57 CLI_OPTION_NO_MSECS => 0,
58 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
59 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_STATISTICS => 0,
62 CLI_OPTION_STRICT_CHECKS => 0,
63 CLI_OPTION_UNBREAK_LINES_ONLY => 0,
64 CLI_OPTION_URL_STATISTICS_THRESHOLD => 0,
65 CLI_OPTION_HOST_STATISTICS_THRESHOLD => 0,
66 CLI_OPTION_SHOW_COMPLETE_REQUEST_DISTRIBUTION => 0,
68 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
70 SHOW_FILTER_READIN_IN => 0,
71 SUPPRESS_EMPTY_LINES => 1,
72 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
73 SUPPRESS_GIF_NOT_CHANGED => 1,
74 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
76 DEBUG_HEADER_REGISTERING => 0,
77 DEBUG_HEADER_HIGHLIGHTING => 0,
80 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
82 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
83 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
85 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
86 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
88 # You better leave these alone unless you know what you're doing.
89 COLOUR_RESET => "\033[0;0m",
93 # For performance reasons, these are global.
96 my %req; # request data from previous lines
101 my $thread_colour_index = 0;
102 my $header_colour_index = 0;
103 my $time_colour_index = 0;
105 my $no_special_header_highlighting;
108 my $header_highlight_regex = '';
110 my $html_output_mode;
112 my $no_msecs_mode; # XXX: should probably be removed
113 my $shorten_thread_ids;
116 sub prepare_our_stuff() {
118 # Syntax Higlight hash
120 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
121 'light_gray', 'light_red', 'light_green', 'yellow',
122 'light_blue', 'pink', 'light_cyan', 'white'
129 Filter => 'purple', # XXX: Used?
130 'Re-Filter' => 'purple',
132 Request => 'light_cyan',
134 CGI => 'light_green',
136 Error => 'light_red',
138 'Fatal error' => 'light_red',
139 'Gif-Deanimate' => 'blue',
141 Writing => 'light_green',
142 Received => 'yellow',
144 # ----------------------
147 request_ => 'brown', # host+path but no protocol
148 'ip-address' => 'yellow',
151 Truncation => 'light_red',
153 Timestamp => 'brown',
154 Crunching => 'light_red',
155 crunched => 'light_red',
156 'Request-Line' => 'pink',
158 destination => 'yellow',
159 'http-version' => 'pink',
160 'crunch-pattern' => 'pink',
165 'program-name' => 'cyan',
168 warning => 'light_red',
169 debug => 'light_red',
173 'status-message' => 'light_cyan',
174 'status-code' => 'yellow',
175 'invalid-request' => 'light_red',
177 error => 'light_red',
178 'rewritten-URL' => 'light_red',
179 'pcrs-delimiter' => 'light_red',
180 'ignored' => 'light_red',
181 'action-bits-update' => 'light_red',
182 'configuration-line' => 'red',
183 'content-type' => 'yellow',
184 'HOST' => HEADER_DEFAULT_COLOUR,
189 # Header colours need their own hash so the keys can be accessed properly
191 # Prefilled with headers that should not appear with default header colours
192 Cookie => 'light_red',
193 'Set-Cookie' => 'light_red',
194 Warning => 'light_red',
195 Default => HEADER_DEFAULT_COLOUR,
198 # Crunch reasons need their own hash as well
200 'Unsupported HTTP feature' => 'light_red',
201 Blocked => 'light_red',
202 Untrusted => 'light_red',
203 Redirected => 'green',
204 'CGI Call' => 'white',
205 'DNS failure' => 'red',
206 'Forwarding failed' => 'light_red',
207 'Connection failure' => 'light_red',
208 'Out of memory (may mask other reasons)' => 'light_red',
209 'No reason recorded' => 'light_red',
212 @time_colours = ('white', 'light_gray');
214 # Translate highlight strings into highlight code
215 prepare_highlight_hash(\%header_colours);
216 prepare_highlight_hash(\%reason_colours);
217 prepare_highlight_hash(\%h);
218 prepare_colour_array(\@all_colours);
219 prepare_colour_array(\@time_colours);
226 ###############################################################
227 # Takes a colour string and returns an ANSI escape sequence
228 # (unless --no-syntax-highlighting is used).
229 # XXX: The Rolling Stones reference has to go.
230 ###############################################################
234 return "" if cli_option_is_set('no-syntax-highlighting');
276 my $bg_code = get_background();
278 our $default = default_colours();
280 if (defined($text{$colour})) {
281 $colour_code = ESCAPE;
282 $colour_code .= $text{$colour};
284 $colour_code .= $light{$colour} ? "1" : "2";
286 $colour_code .= $bg_code;
288 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
290 } elsif ($colour =~ /reset/) {
292 $colour_code = default_colours();
296 die "What's $colour supposed to mean?\n";
302 sub get_semantic_html_markup($) {
303 ###############################################################
304 # Takes a string and returns a span element
305 ###############################################################
310 if ($type =~ /Standard/) {
314 $code = '<span title="' . $type . '" class="' . $type . '">';
320 sub cli_option_is_set($) {
323 my $cli_option = shift;
325 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
327 return $cli_options{$cli_option};
330 sub get_html_title() {
333 return $cli_options{'title'};
337 sub init_css_colours() {
345 purple => "F06", # XXX: wrong
346 cyan => "F09", # XXX: wrong
351 light_green => "33F",
360 sub get_css_colour($) {
365 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
367 return '#' . $css_colours{$colour};
370 sub get_css_line($) {
375 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
376 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
377 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
378 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
379 $css_line .= '}' . "\n";
384 sub get_css_line_for_colour($) {
389 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
390 $css_line .= 'color:' . get_css_colour($colour) . ';';
391 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
392 $css_line .= '}' . "\n";
397 # XXX: Wrong solution
398 sub get_missing_css_lines() {
402 $css_line .= '.' . 'default' . ' {';
403 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
404 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
405 $css_line .= '}' . "\n";
412 our %css_colours; #XXX: Wrong solution
416 $css .= '.privoxy-log {';
417 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
418 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
421 foreach my $key (keys %h_colours) {
423 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
424 $css .= get_css_line($key);
428 foreach my $colour (keys %css_colours) {
430 $css .= get_css_line_for_colour($colour);
434 $css .= get_missing_css_lines(); #XXX: Wrong solution
443 if (cli_option_is_set('html-output')) {
445 my $title = get_html_title();
447 $intro .= '<html><head>';
448 $intro .= '<title>' . $title . '</title>';
449 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
450 $intro .= '</head><body>';
451 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
461 if (cli_option_is_set('html-output')) {
463 $outro = '</p></body></html>';
470 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
473 sub get_colour_html_markup($) {
474 ###############################################################
475 # Takes a colour string a span element. XXX: WHAT?
476 # XXX: This function shouldn't be necessary, the
477 # markup should always be semantically correct.
478 ###############################################################
483 if ($type =~ /Standard/) {
486 $code = '<span class="' . lc($type) . '">';
492 sub default_colours() {
495 return reset_colours();
502 sub reset_colours() {
503 return ESCAPE . "0m";
506 sub set_background($) {
522 if (defined($backgrounds{$colour})) {
523 $bg_code = $backgrounds{$colour};
525 die "Invalid background colour: " . $colour;
529 sub get_background() {
533 sub prepare_highlight_hash($) {
536 foreach my $key (keys %$ref) {
537 $$ref{$key} = $html_output_mode ?
538 get_semantic_html_markup($key) :
539 paint_it($$ref{$key});
543 sub prepare_colour_array($) {
546 foreach my $i (0 ... @$ref - 1) {
547 $$ref[$i] = $html_output_mode ?
548 get_colour_html_markup($$ref[$i]) :
553 sub found_unknown_content($) {
558 return unless cli_option_is_set('strict-checks');
560 return if ($unknown =~ /\[too long, truncated\]$/);
562 $message = "found_unknown_content: Don't know how to highlight: ";
563 # Break line so the log file can later be parsed as Privoxy log file again
564 $message .= '"' . $unknown . '"' . " in:\n";
565 $message .= $req{$t}{'log-message'};
566 debug_message($message);
567 log_parse_error($req{$t}{'log-message'});
569 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
572 sub log_parse_error($) {
576 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
577 open(my $errorlog_fd, ">>", ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
578 print $errorlog_fd $message;
583 sub debug_message(@) {
586 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
589 ################################################################################
590 # highlighter functions that aren't loglevel-specific
591 ################################################################################
595 # Get highlight marker
596 my $highlight = shift; # XXX: Stupid name;
600 if (defined($highlight)) {
602 $result = $h{$highlight};
606 $message = "h: Don't recognize highlighter $highlight.";
607 debug_message($message);
608 log_parser_error($message);
609 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
615 sub highlight_known_headers($) {
619 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
621 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
623 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
624 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
630 sub highlight_matched_request_line($$) {
632 my $result = shift; # XXX: Stupid name;
634 if ($result =~ m@(.*)($regex)(.*)@) {
635 $result = $1 . highlight_request_line($2) . $3
640 sub highlight_request_line($) {
643 my ($method, $url, $http_version);
645 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
646 if ($rl =~ m/Invalid request/) {
648 $rl = h('invalid-request') . $rl . h('Standard');
650 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d+\.\d+)/) {
652 # XXX: might not match in case of HTTP method fuzzing.
653 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
654 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
655 if ($rl =~ /http:\/\//) {
656 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
658 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
661 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
663 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
665 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
666 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
668 } elsif ($rl =~ m/^ $/) {
670 $rl = h('error') . "No request line specified!" . h('Standard');
674 debug_message ("Can't parse request line: $rl");
681 sub highlight_response_line($) {
684 my ($http_version, $status_code, $status_message);
689 # TODO: Mark different status codes differently
691 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
692 ($http_version, $status_code, $status_message) = ($1, $2, $3);
694 debug_message ("Can't parse response line: $rl") and die 'Fix this';
697 # Rebuild highlighted
699 $rl .= h('http-version') . $http_version . h('Standard');
701 $rl .= h('status-code') . $status_code . h('Standard');
703 $rl .= h('status-message') . $status_message . h('Standard');
708 sub highlight_matched_url($$) {
710 my $result = shift; # XXX: Stupid name;
713 #print "Got $result, regex ($regex)\n";
715 if ($result =~ m@(.*?)($regex)(.*)@) {
716 $result = $1 . highlight_url($2) . $3;
717 #print "Now the result is $result\n";
723 sub highlight_matched_host($$) {
725 my ($result, $regex) = @_; # XXX: result ist stupid name;
727 if ($result =~ m@(.*?)($regex)(.*)@) {
728 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
734 sub highlight_matched_pattern($$$) {
736 my $result = shift; # XXX: Stupid name;
740 die "Unknown key $key" unless defined $h{$key};
742 if ($result =~ m@(.*?)($regex)(.*)@) {
743 $result = $1 . h($key) . $2 . h('Standard') . $3;
749 sub highlight_matched_path($$) {
751 my $result = shift; # XXX: Stupid name;
754 if ($result =~ m@(.*?)($regex)(.*)@) {
755 $result = $1 . h('path') . $2 . h('Standard') . $3;
761 sub highlight_url($) {
765 if ($html_output_mode) {
767 $url = '<a href="' . $url . '">' . $url . '</a>';
771 $url = h('URL') . $url . h('Standard');
778 sub update_header_highlight_regex($) {
781 my $headers = join ('|', keys %header_colours);
783 $header_highlight_regex = qr/$headers/;
784 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
787 ################################################################################
788 # loglevel-specific highlighter functions
789 ################################################################################
791 sub handle_loglevel_header($) {
795 if ($c =~ /^scan:/) {
797 if ($c =~ m/^scan: ([^: ]+):/) {
799 # Register new headers
800 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
802 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
803 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
805 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
806 $header_colours{$header} = $header_colours{'Default'};
808 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
809 $header_colour_index++;
811 update_header_highlight_regex($header);
814 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
816 # scan: GET http://p.p/ HTTP/1.1
817 $c = $1 . highlight_request_line($2);
819 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
821 # scan: HTTP/1.1 200 OK
822 $req{$t}{'response_line'} = $2;
823 $req{$t}{'status_code'} = $3;
824 $req{$t}{'status_message'} = $4;
825 $c = $1 . highlight_response_line($req{$t}{'response_line'});
828 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
830 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
831 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
832 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
834 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
836 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
837 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
838 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
840 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
842 # Text mode enabled by force. Take cover!
843 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
845 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
847 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
848 $c = $1 . highlight_request_line($2);
850 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
852 # Adjusted Content-Length to 2132
853 # Adjust Content-Length to 33533
854 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
855 $c = highlight_known_headers($c);
857 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
859 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
860 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
862 } elsif ($c =~ m/^Couldn\'t parse:/) {
864 # XXX: These should probable be logged with LOG_LEVEL_ERROR
865 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
866 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
867 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
869 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
870 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
872 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
873 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
874 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
876 # XXX: Save tag and tagger
878 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
879 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
880 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
881 $no_special_header_highlighting = 1;
883 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
885 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
886 # XXX: Save tag and tagger
888 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
889 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
891 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
892 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
893 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
894 or $c =~ m/^(addh-unique|Referer forged to)/
895 or $c =~ m/^Downgraded answer to HTTP\/1.0/
896 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
897 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
898 or $c =~ m/^Referer crunched!/
899 or $c =~ m/^crunched x-forwarded-for!/
900 or $c =~ m/^crunched From!/
901 or $c =~ m/^ modified$/
902 or $c =~ m/^Content filtering is enabled. Crunching:/
903 or $c =~ m/^force-text-mode overruled the client/
904 or $c =~ m/^Server time in the future\./
905 or $c =~ m/^content-disposition header crunched and replaced with:/i
906 or $c =~ m/^Reducing white space in /
907 or $c =~ m/^Ignoring single quote in /
908 or $c =~ m/^Converting tab to space in /
909 or $c =~ m/A HTTP\/1\.1 response without/
910 or $c =~ m/Disabled filter mode on behalf of the client/
911 or $c =~ m/Keeping the (?:server|client) header /
912 or $c =~ m/Content modified with no Content-Length header set/
913 or $c =~ m/^Appended client IP address to/
914 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
915 or $c =~ m/^keep-alive support is disabled/
916 or $c =~ m/^Continue hack in da house/
917 or $c =~ m/^Merged multiple header lines to:/
918 or $c =~ m/^Added header: /
919 or $c =~ m/^Enlisting (?:sorted|left-over) header/
920 or $c =~ m/^Multiple Content-Type headers detected. Removing and ignoring: Content-Type:/
923 # XXX: Some of these may need highlighting
925 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
926 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
927 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
928 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
929 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
930 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
931 # Text mode is already enabled
932 # Denied request with NULL byte(s) turned into line break(s)
933 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
934 # addh-unique: Host: people.freebsd.org
935 # Suppressed offer to compress content
936 # Crunched incoming cookie -- yum!
937 # Accepted the client's request to fetch without filtering.
938 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
939 # addh-unique: Host: subkeys.pgp.net:11371
940 # Referer forged to: Referer: http://10.0.0.1/
941 # Downgraded answer to HTTP/1.0
942 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
943 # Referer overwritten with: Referer: pille-palle
944 # Referer replaced with: Referer: pille-palle
945 # crunched x-forwarded-for!
947 # modified # XXX: pretty stupid log message
948 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
949 # force-text-mode overruled the client's request to fetch without filtering!
950 # Server time in the future.
951 # content-disposition header crunched and replaced with: content-disposition: filename=baz
952 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
953 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
954 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
955 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
957 # A HTTP/1.1 response without Connection header implies keep-alive.
958 # Disabled filter mode on behalf of the client.
959 # Keeping the server header 'Connection: keep-alive' around.
960 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
961 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
962 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
963 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
964 # Removing 'Connection: close' to imply keep-alive.
965 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
966 # Continue hack in da house.
967 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
968 # Added header: Content-Encoding: deflate
969 # Enlisting sorted header User-Agent: Mozilla/5.0 (X11; SunOS i86pc; rv:10.0.3) Gecko/20100101 Firefox/10.0.3
970 # Enlisting left-over header Connection: close
971 # Multiple Content-Type headers detected. Removing and ignoring: Content-Type: text/html
973 } elsif ($c =~ m/^scanning headers for:/) {
975 return '' unless SHOW_SCAN_INTRO;
977 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
978 # crunched User-Agent!
979 # Crunching: Content-Encoding: gzip
981 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
983 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
985 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
987 $c = h('warning') . $c . h('Standard');
989 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
991 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
992 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
994 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
995 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
997 } elsif ($c =~ m/^Removing empty header/) {
999 # Removing empty header
1002 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
1004 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
1005 # Enable force-text-mode if you know what you're doing.
1006 # XXX: Could highlight more here.
1007 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1009 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1011 # Server keep-alive timeout is 5. Sticking with 10.
1012 # Client keep-alive timeout is 20. Sticking with 10.
1014 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1015 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1017 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1019 # Reducing keep-alive timeout from 60 to 10.
1021 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1022 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1024 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1026 # Killed all-caps Host header line: HOST: bestproxydb.com
1027 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1028 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1032 found_unknown_content($c);
1036 unless ($c =~ m/^Transforming/) {
1037 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1043 sub handle_loglevel_re_filter($) {
1045 my $content = shift;
1049 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1051 # XXX: only the second version gets highlighted properly.
1052 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1053 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1054 $req{$t}{'content_source'} = $1;
1055 $req{$t}{'content_size'} = $2;
1056 $req{$t}{'content_filter'} = $3;
1057 $req{$t}{'content_hits'} = $4;
1058 $req{$t}{'new_content_size'} = $5;
1059 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1060 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1061 if ($req{$t}{'content_hits'} == 0 and
1062 not (cli_option_is_set('show-ineffective-filters')
1063 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1067 $c =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1068 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1069 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1071 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1072 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1075 $c .= "(" . $h{'Number'};
1076 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1077 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1080 } elsif ($c =~ /\.{3}$/
1081 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1083 # Used by Privoxy 3.0.5 and 3.0.6:
1085 # Used by Privoxy 3.0.7:
1086 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1088 $req{$t}{'filtered_header'} = $1;
1089 $req{$t}{'old_header_size'} = $2;
1090 $req{$t}{'header_filter_name'} = $3;
1092 unless (cli_option_is_set('show-ineffective-filters') or
1093 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1096 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1097 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1099 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1101 # ...produced 0 hits (new size 23).
1102 #... produced 1 hits (new size 54).
1104 $req{$t}{'header_filter_hits'} = $1;
1105 $req{$t}{'new_header_size'} = $2;
1107 unless (cli_option_is_set('show-ineffective-filters') or
1108 (defined($req{$t}{'header_filter_name'}) and
1109 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1111 if ($req{$t}{'header_filter_hits'} == 0 and
1112 not (defined($req{$t}{'header_filter_name'}) and
1113 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1116 # Reformat including information from the intro
1117 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1119 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1120 # $c .= $req{$t}{'filtered_header'} . " ";
1121 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1122 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1124 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1126 $c .= "changing size from ";
1127 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1129 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1134 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1138 # Highlight from last line (XXX: What?)
1139 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1140 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1145 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1146 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1151 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1153 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1154 # Tagger variable-test has empty joblist. Nothing to do.
1156 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1158 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1160 $req{$t}{'chunked-size'} = $1;
1161 $req{$t}{'dechunked-size'} = $2;
1162 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1164 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1166 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1167 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1169 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1171 # Decompression successful. Old size: 670, new size: 1166.
1173 $req{$t}{'size-compressed'} = $1;
1174 $req{$t}{'size-decompressed'} = $2;
1175 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1177 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1178 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1180 # XXX: Create sub get_percentage()
1181 if ($req{$t}{'size-decompressed'}) {
1182 $req{$t}{'decompression-gain-percent'} =
1183 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1185 $content .= " (saved: ";
1186 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1188 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1192 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1194 # Need to de-chunk first
1195 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1197 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1199 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1201 # Adding re_filter job ...
1202 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1203 # to filter client-ip-address succeeded.
1205 } elsif ($c =~ m/^Compressed content from /) {
1207 # Compressed content from 29258 to 8630 bytes. Compression level: 3
1208 $content =~ s@(?<=from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1209 $content =~ s@(?<=to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1210 $content =~ s@(?<=level: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1212 } elsif ($c =~ m/^Reading in filter/) {
1214 return '' unless SHOW_FILTER_READIN_IN;
1216 } elsif ($c =~ m/^Decompression didn't result/) {
1218 # Decompression didn't result in any content.
1220 # Nothing to highlight.
1224 found_unknown_content($content);
1231 sub handle_loglevel_tagging($) {
1235 if ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
1236 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
1238 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
1239 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
1240 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
1242 # XXX: Save tag and tagger
1244 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
1245 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
1246 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
1251 sub handle_loglevel_redirect($) {
1255 if ($c =~ m/^Decoding "([^""]*)"/) {
1257 $req{$t}{'original-destination'} = $1;
1258 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1261 } elsif ($c =~ m/^Checking/) {
1263 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1264 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1266 # TODO: Change colour if really url-decoded
1267 $req{$t}{'decoded-original-destination'} = $1;
1268 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1271 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1273 # pcrs command "s@&from=rss@@" changed \
1274 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1275 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1276 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1277 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1278 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1279 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1280 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1281 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1283 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1285 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1286 # change "http://www.example.org/".
1287 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1288 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1290 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1292 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1293 # XXX: Use URL highlighter
1295 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1297 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1298 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1299 # is already properly formatted.
1300 # XXX: assume the same?
1301 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1303 } elsif ($c =~ m/^Percent-encoding redirect/) {
1305 # Percent-encoding redirect URL: http://www.example.org/\x02
1306 $c = highlight_matched_url($c, '(?<=redirect URL: ).*');
1308 } elsif ($c =~ m/^Rewrite detected:/) {
1310 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1311 # Rewrite detected: GET https://www.electrobsd.org/CommonJS/ajax/libs/jquery/3.4.1/jquery.min.js HTTP/1.1
1312 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1316 found_unknown_content($c);
1323 sub handle_loglevel_gif_deanimate($) {
1325 my $content = shift;
1327 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1329 my $bytes_from = $1;
1331 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1332 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1333 # XXX: Do we need g in case of ($1 == $2)?
1334 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1336 } elsif ($content =~ m/GIF (not) changed/) {
1338 # Gif-Deanimate: GIF not changed.
1339 return '' if SUPPRESS_GIF_NOT_CHANGED;
1340 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1342 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1344 # failed! (gif parsing)
1345 # XXX: Replace this error message with something less stupid
1346 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1348 } elsif ($content =~ m/^Need to de-chunk first/) {
1350 # Need to de-chunk first
1351 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1353 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1355 # No GIF header found (XXX: Did I ever commit this?)
1356 # failed while parsing 195 134747048 (XXX: never committed)
1358 # Ignore these for now
1362 found_unknown_content($content);
1369 sub handle_loglevel_request($) {
1371 my $content = shift;
1373 if ($content =~ m/crunch! /) {
1375 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1377 # Highlight crunch reasons
1378 foreach my $reason (keys %reason_colours) {
1379 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1381 # Highlight request URL domain and ditch 'crunch!'
1382 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1383 $content =~ s@ crunch!@@;
1385 } elsif ($content =~ m/\[too long, truncated\]$/) {
1387 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1388 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1390 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1392 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1393 $content = h('request_') . $content . h('Standard');
1397 found_unknown_content($content);
1404 sub handle_loglevel_crunch($) {
1406 my $content = shift;
1408 # Highlight crunch reason
1409 foreach my $reason (keys %reason_colours) {
1410 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1413 if ($content =~ m/\[too long, truncated\]$/) {
1415 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1416 # [...]&filter... [too long, truncated]
1417 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1421 # Blocked: http://ads.example.org/
1422 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1428 sub handle_loglevel_connect($) {
1432 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1434 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1436 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1437 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1439 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1441 # connect to: www.example.org.noconnect failed: Operation not permitted
1443 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1445 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1447 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1449 # Connect: to www.nzherald.co.nz successful
1450 # Connect: to archiv.radiotux.de
1452 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1453 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1455 } elsif ($c =~ m/^to ([^\s]*)$/) {
1457 # Connect: to lists.sourceforge.net:443
1459 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1461 } elsif ($c =~ m/^[Aa]ccepted connection from .*/ or
1465 # Accepted connection from 10.0.0.1 on socket 5
1466 # Privoxy between 3.0.20 and 3.0.6:
1467 # accepted connection from 10.0.0.1( on socket 5)?
1468 # Privoxy 3.0.6 and earlier just say:
1470 $c = highlight_matched_host($c, '(?<=connection from )[^ ]*');
1471 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1473 } elsif ($c =~ m/^Closing client socket/) {
1475 # Closing client socket 5. Keep-alive: 0, Socket alive: 1. Data available: 0.
1476 # Privoxy 3.0.20 and later
1477 # Closing client socket 8. Keep-alive: 1. Socket alive: 0. Data available: 0. \
1478 # Configuration file change detected: 0. Requests received: 11.
1480 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1481 $c = highlight_matched_pattern($c, 'Number', '(?<=Keep-alive: )\d+');
1482 $c = highlight_matched_pattern($c, 'Number', '(?<=Socket alive: )\d+');
1483 $c = highlight_matched_pattern($c, 'Number', '(?<=available: )\d+');
1484 $c = highlight_matched_pattern($c, 'Number', '(?<=detected: )\d+');
1485 $c = highlight_matched_pattern($c, 'Number', '(?<=received: )\d+');
1487 } elsif ($c =~ m/^write header to: .* failed:/) {
1489 # write header to: 10.0.0.1 failed: Broken pipe
1491 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1492 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1494 } elsif ($c =~ m/^write header to client failed:/) {
1496 # write header to client failed: Broken pipe
1498 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1500 } elsif ($c =~ m/^socks4_connect:/) {
1502 # socks4_connect: SOCKS request rejected or failed.
1503 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1505 } elsif ($c =~ m/^Listening for new connections/ or
1506 $c =~ m/^accept connection/) {
1508 # Privoxy versions above 3.0.6 say:
1509 # Listening for new connections ...
1510 # earlier versions say:
1511 # accept connection ...
1514 } elsif ($c =~ m/^accept failed:/) {
1516 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1518 } elsif ($c =~ m/^Overriding forwarding settings/) {
1520 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1521 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1523 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1525 # Denying suspicious CONNECT request from 10.0.0.1
1526 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1528 } elsif ($c =~ m/^socks5_connect:/) {
1530 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1532 } elsif ($c =~ m/^Created new connection to/) {
1534 # Created new connection to www.privoxy.org:80 on socket 11.
1535 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1536 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1538 } elsif ($c =~ m/^Found reusable socket/) {
1540 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1542 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1543 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1544 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1545 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1546 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1547 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1548 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1549 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1551 } elsif ($c =~ m/^Marking open socket/) {
1553 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1554 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1555 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1556 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 } elsif ($c =~ m/^No reusable/) {
1560 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1561 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1563 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1565 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1566 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1568 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1569 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1570 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1572 } elsif ($c =~ m/^Socket/) {
1574 # Socket 16 already forgotten or never remembered.
1575 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1577 } elsif ($c =~ m/^The connection to/) {
1579 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1581 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1582 # Timeout is: 1. Assumed latency: 4.
1583 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1584 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1585 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1586 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1587 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1588 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1590 } elsif ($c =~ m/^Stopped waiting for the request line/ or
1591 $c =~ m/^No request line on socket \d received in time/ or
1592 $c =~ m/^The client side of the connection on socket \d/) {
1594 # Stopped waiting for the request line. Timeout: 121.
1595 # Privoxy 3.0.19 and later:
1596 # No request line on socket 5 received in time. Timeout: 1.
1597 # The client side of the connection on socket 5 got closed \
1598 # without sending a complete request line.
1599 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1600 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1602 } elsif ($c =~ m/^Waiting for \d/) {
1604 # Waiting for 1 connections to timeout.
1605 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 } elsif ($c =~ m/^Initialized/) {
1609 # Initialized 20 socket slots.
1610 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1612 } elsif ($c =~ m/^Done reading from server/) {
1614 # Done reading from server. Expected content length: 24892. \
1615 # Actual content length: 24892. Most recently received: 4412.
1617 # Done reading from server. Expected content length: 24892. \
1618 # Actual content length: 24892. Bytes most recently read: 4412.
1619 # Done reading from server. Content length: 6018 as expected. \
1620 # Bytes most recently read: 294.
1621 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1622 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1623 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1625 } elsif ($c =~ m/^Continuing buffering (?:server )?headers/) {
1627 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1628 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1629 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1630 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1631 # 3.0.15 up to 3.0.19:
1632 # Continuing buffering headers. Bytes most recently read: 498.
1633 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1635 # Continuing buffering server headers from socket 5. Bytes most recently read: 498.
1636 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1638 } elsif ($c =~ m/^Received \d+ bytes while/) {
1640 # Received 206 bytes while expecting 12103.
1641 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1642 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1644 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1646 # Connection from 81.163.28.218 dropped due to ACL
1647 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1648 # Connection from 192.168.2.1 on 127.0.1.1:8118 (socket 3) dropped due to ACL
1649 $c = highlight_matched_host($c, '(?<=onnection from )[\d.:]+');
1650 $c = highlight_matched_host($c, '(?<=on )[\d.:]+');
1651 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1653 } elsif ($c =~ m/^(?:Reusing|Closing) server socket / or
1654 $c =~ m/^No additional client request/) {
1656 # Reusing server socket 4. Opened for 10.0.0.1.
1657 # Closing server socket 2. Opened for 10.0.0.1.
1658 # No additional client request received in time. \
1659 # Closing server socket 4, initially opened for 10.0.0.1.
1660 # No additional client request received in time on socket 29.
1661 # Privoxy 3.0.20 and later
1662 # Reusing server socket 7 connected to www.privoxy.org. Total requests: 2.
1663 # Closing server socket 6 connected to d.asset.soup.io. Keep-alive: 0.\
1664 # Tainted: 1. Socket alive: 1. Timeout: 60. Configuration file change detected: 0.
1665 # Reusing server socket 35 connected to nl.wikipedia.org. Requests already sent: 5.
1667 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1668 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.)');
1669 $c = highlight_matched_host($c, '(?<=connected to )[^\s]+(?=\.)');
1670 for my $number_pattern ('requests', 'Keep-alive', 'Tainted', ' alive', 'Timeout', 'detected') {
1671 $c = highlight_matched_pattern($c, 'Number', '(?<='. $number_pattern . ': )\d+');
1673 $c =~ s@(?<=already sent: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1675 } elsif ($c =~ m/^Connected to /) {
1677 # Connected to tor-jail[10.0.0.2]:9050.
1679 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1680 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1681 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1683 } elsif ($c =~ m/^Could not connect to /) {
1685 # Could not connect to [10.0.0.1]:80.
1687 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1688 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1690 } elsif ($c =~ m/^Waiting for the next client request/ or
1691 $c =~ m/^The connection on server socket/ or
1692 $c =~ m/^Client request (?:\d+ )?(?:arrived in time|has been pipelined) /) {
1694 # Waiting for the next client request on socket 3. Keeping the server \
1695 # socket 12 to a.fsdn.com open.
1696 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1697 # Privoxy 3.0.20 and later:
1698 # Client request 4 arrived in time on socket 7.
1699 # Used by Privoxy 3.0.18 and 3.0.19:
1700 # Client request arrived in time on socket 21.
1701 # Used by earlier version:
1702 # Client request arrived in time or the client closed the connection on socket 12.
1703 # Client request 8 has been pipelined on socket 7 and the socket is still alive.
1705 $c =~ s@(?<=request )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1706 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1707 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1708 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1710 } elsif ($c =~ m/^Marking the server socket/) {
1712 # Marking the server socket 7 tainted.
1714 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1716 } elsif ($c =~ m/^Reduced expected bytes to /) {
1718 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1719 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1720 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1722 } elsif ($c =~ m/^The client closed socket /) {
1724 # The client closed socket 2 while the server socket 4 is still open.
1725 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1726 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1728 } elsif ($c =~ m/^Expected client content length set /) {
1730 # Expected client content length set to 667325411 after reading 4999 bytes.
1731 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1732 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1734 } elsif ($c =~ m/^Reducing expected bytes to /) {
1736 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1737 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1738 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1740 } elsif ($c =~ m/^Waiting for up to /) {
1742 # Waiting for up to 4999 bytes from the client.
1743 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1745 } elsif ($c =~ m/^Optimistically sending /) {
1747 # Optimistically sending 318 bytes of client headers intended for www.privoxy.org
1748 $c =~ s@(?<=sending )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1749 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1751 } elsif ($c =~ m/^Stopping to watch the client socket/) {
1753 # Stopping to watch the client socket. There's already another request waiting.
1754 # Privoxy 3.0.20 and later:
1755 # Stopping to watch the client socket 5. There's already another request waiting.
1756 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1758 } elsif ($c =~ m/^Drained \d+ bytes before closing/) {
1760 # Drained 180 bytes before closing socket 6
1761 $c =~ s@(?<=Drained )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1762 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1764 } elsif ($c =~ m/^Tainting client socket/ or
1765 $c =~ m/^Failed to shutdown socket/) {
1767 # Tainting client socket 7 due to unread data.
1768 # Failed to shutdown socket 11: Connection reset by peer
1770 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1772 } elsif ($c =~ m/^Shifting \d+ pipelined bytes/) {
1774 # Shifting 360 pipelined bytes by 360 bytes
1775 $c =~ s@(?<=Shifting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1776 $c =~ s@(?<=by )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1778 } elsif ($c =~ m/^Flushed (\d+) bytes of request body while expecting (\d+)/) {
1780 # Flushed 30 bytes of request body while expecting 30
1781 $c =~ s@(?<=Flushed )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1782 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1784 } elsif ($c =~ m/^Performing the TLS\/SSL handshake with client. Hash of host:/) {
1786 # Performing the TLS/SSL handshake with client. Hash of host: bab5296b25e256c7b06b92b17b56bcae
1787 $c = highlight_matched_host($c, '(?<=Hash of host: ).+');
1789 } elsif ($c =~ m/^Forwarding \d+ bytes of encrypted POST data/) {
1791 # Forwarding 1954 bytes of encrypted POST data
1792 $c =~ s@(?<=Forwarding )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1794 } elsif ($c =~ m/^Forwarded the last \d+ bytes/) {
1796 # Forwarded the last 1954 bytes
1797 $c =~ s@(?<=the last )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1799 } elsif ($c =~ m/^Waiting for the next client connection. Currently active threads:/) {
1801 # Waiting for the next client connection. Currently active threads: 30
1802 $c =~ s@(?<=threads: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1804 } elsif ($c =~ m/^Data arrived in time on client socket/) {
1806 # Data arrived in time on client socket 6. Requests so far: 3
1807 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1808 $c =~ s@(?<=Requests so far: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1810 } elsif ($c =~ m/^Dropping the client connection on socket/) {
1812 # Dropping the client connection on socket 71. The server connection has not been established yet.
1813 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1815 } elsif ($c =~ m/^The client socket \d+ has become unusable while the server/) {
1817 # The client socket 16 has become unusable while the server socket 24 is still open.
1818 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1819 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1821 } elsif ($c =~ m/^Looks like we / or
1822 $c =~ m/^Unsetting keep-alive flag/ or
1823 $c =~ m/^No connections to wait/ or
1824 $c =~ m/^Complete client request received/ or
1825 $c =~ m/^Possible pipeline attempt detected./ or
1826 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1827 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1828 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1829 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1830 $c =~ m/^There better be a request body./ or
1831 $c =~ m/^Done reading from the client\.$/) {
1833 # Looks like we reached the end of the last chunk. We better stop reading.
1834 # Looks like we read the end of the last chunk together with the server \
1835 # headers. We better stop reading.
1836 # Looks like we got the last chunk together with the server headers. \
1837 # We better stop reading.
1838 # Unsetting keep-alive flag.
1839 # No connections to wait for left.
1840 # Client request arrived in time or the client closed the connection.
1841 # Complete client request received
1842 # Possible pipeline attempt detected. The connection will not be \
1843 # kept alive and we will only serve the first request.
1844 # POST request detected. The connection will not be kept alive.
1845 # The server still wants to talk, but the client hung up on us.
1846 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1847 # There might be a request body. The connection will not be kept alive.
1848 # Privoxy 3.0.20 and later
1849 # There better be a request body.
1850 # Done reading from the client.
1854 found_unknown_content($c);
1862 sub handle_loglevel_info($) {
1866 if ($c =~ m/^Rewrite detected:/) {
1868 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1869 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1871 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1872 $c =~ m/^Compressed content detected/ or
1873 $c =~ m/^SDCH-compressed content detected/ or
1876 # Decompressing deflated iob: 117
1877 # Decompression didn't result in any content.
1878 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1879 # with zlib support or enable the prevent-compression action.
1880 # SDCH-compressed content detected, content filtering disabled.\
1881 # Consider suppressing SDCH offers made by the client.
1882 # Tagger 'complete-url' created empty tag. Ignored.
1886 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1888 # loading configuration file '/usr/local/etc/privoxy/config':
1889 # Reloading configuration file '/usr/local/etc/privoxy/config'
1890 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1892 } elsif ($c =~ m/^Loading (actions|filter|trust) file: /) {
1894 # Loading actions file: /usr/local/etc/privoxy/default.action
1895 # Loading filter file: /usr/local/etc/privoxy/default.filter
1896 # Loading trust file: /usr/local/etc/privoxy/trust
1898 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1900 } elsif ($c =~ m/^exiting by signal/) {
1902 # exiting by signal 15 .. bye
1903 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1905 } elsif ($c =~ m/^Privoxy version/) {
1907 # Privoxy version 3.0.7
1908 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1910 } elsif ($c =~ m/^Program name: /) {
1912 # Program name: /usr/local/sbin/privoxy
1913 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1915 } elsif ($c =~ m/^Listening on port /) {
1917 # Listening on port 8118 on IP address 10.0.0.1
1918 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1919 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1921 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1923 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1924 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1926 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1928 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1929 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1931 # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
1932 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1934 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1935 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1936 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1937 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1938 $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
1940 } elsif ($c =~ m/^Status code/) {
1942 # Status code 304 implies no body.
1943 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1945 } elsif ($c =~ m/^Method/) {
1947 # Method HEAD implies no body.
1948 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1950 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1952 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1953 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1954 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1956 } elsif ($c =~ m/^File modification detected: /) {
1958 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1959 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1961 } elsif ($c =~ m/^No logfile configured/ or
1962 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1963 $c =~ m/^Invalid \"chunked\" transfer/ or
1964 $c =~ m/^Support for/ or
1965 $c =~ m/^Flushing header and buffers/ or
1966 $c =~ m/^Can not resolve/
1969 # No logfile configured. Please enable it before reporting any problems.
1970 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1971 # response or even no response at all.
1972 # No logfile configured. Logging disabled.
1973 # Invalid "chunked" transfer encoding detected and ignored.
1974 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1975 # known not to work properly in some situations.
1976 # Flushing header and buffers. Stepping back from filtering.
1977 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1981 found_unknown_content($c);
1988 sub handle_loglevel_cgi($) {
1992 if ($c =~ m/^Granting access to/) {
1994 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1996 } elsif ($c =~ m/^Substituting: s(.)/) {
1998 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1999 # XXX: prone to span several lines
2002 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
2003 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
2004 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
2010 sub handle_loglevel_force($) {
2014 if ($c =~ m/^Ignored force prefix in request:/) {
2016 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
2017 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
2018 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
2020 } elsif ($c =~ m/^Enforcing request:/) {
2022 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
2023 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
2027 found_unknown_content($c);
2034 sub handle_loglevel_error($) {
2038 if ($c =~ m/^(?:Empty|No) server or forwarder response received on socket \d+\./) {
2040 # Empty server or forwarder response received on socket 4.
2041 # Empty server or forwarder response received on socket 3. \
2042 # Closing client socket 15 without sending data.
2043 # Used by Privoxy 3.0.18 and later:
2044 # No server or forwarder response received on socket 8. \
2045 # Closing client socket 10 without sending data.
2047 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
2048 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
2050 } elsif ($c =~ m/^Didn't receive data in time:/) {
2052 # Didn't receive data in time: a.fsdn.com:443
2053 $c =~ s@(?<=in time: )(.*)@$h{'destination'}$1$h{'Standard'}@;
2056 # XXX: There are probably more messages that deserve highlighting.
2062 sub handle_loglevel_ignore($) {
2066 sub gather_loglevel_clf_stats($) {
2068 my $content = shift;
2069 my ($method, $resource, $http_version, $status_code, $size);
2073 # +0200] "GET https://www.youtube.com/watch?v=JmcA9LIIXWw HTTP/1.1" 200 68004
2074 # +0200] "VERSION-CONTROL http://p.p/ HTTP/1.1" 200 2787
2075 $content =~ m/^[+-]\d{4}\] "([^ ]+) (.+) (HTTP\/\d\.\d)" (\d+) (\d+)/;
2082 $stats{requests_clf}++;
2084 unless (defined $method) {
2085 # +0200] "Invalid request" 400 0
2086 return if ($content =~ m/^[+-]\d{4}\] "Invalid request"/);
2087 print("Failed to parse: $content\n");
2090 $stats{'method'}{$method}++;
2091 if ($cli_options{'url-statistics-threshold'} != 0) {
2092 $stats{'resource'}{$resource}++;
2094 $stats{'http-version'}{$http_version}++;
2096 if ($cli_options{'host-statistics-threshold'} != 0) {
2097 $resource =~ m@(?:https?://)?([^/]+)/?@;
2098 $stats{'hosts'}{$1}++;
2100 $stats{'content-size-total'} += $size;
2101 $stats{'status-code'}{$status_code}++;
2104 sub gather_loglevel_request_stats($$) {
2112 sub gather_loglevel_crunch_stats($$) {
2119 if ($c =~ m/^Redirected:/) {
2120 # Redirected: http://www.example.org/http://p.p/
2121 $stats{'fast-redirections'}++;
2123 } elsif ($c =~ m/^Blocked:/) {
2124 # Blocked: blogger.googleusercontent.com:443
2125 $stats{'blocked'}++;
2127 } elsif ($c =~ m/^Connection timeout:/) {
2128 # Connection timeout: http://c.tile.openstreetmap.org/18/136116/87842.png
2129 $stats{'connection-timeout'}++;
2131 } elsif ($c =~ m/^Connection failure:/) {
2132 # Connection failure: http://127.0.0.1:8080/
2133 $stats{'connection-failure'}++;
2138 sub gather_loglevel_error_stats($$) {
2145 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
2147 # Empty server or forwarder response received on socket 4.
2148 $stats{'empty-responses'}++;
2149 if ($thread_data{$thread}{'new_connection'}) {
2150 $stats{'empty-responses-on-new-connections'}++;
2152 $stats{'empty-responses-on-reused-connections'}++;
2157 sub gather_loglevel_connect_stats($$) {
2159 my ($c, $thread) = @_;
2163 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
2165 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
2166 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
2168 } elsif ($c =~ m/^to ([^\s]*)$/) {
2170 # Connect: to lists.sourceforge.net:443
2172 $thread_data{$thread}{'forwarder'} = 'direct connection';
2174 } elsif ($c =~ m/^Created new connection to/) {
2176 # Created new connection to www.privoxy.org:80 on socket 11.
2178 $thread_data{$thread}{'new_connection'} = 1;
2180 } elsif ($c =~ m/^Reusing server socket \d./ or
2181 $c =~ m/^Found reusable socket/) {
2183 # Reusing server socket 4. Opened for 10.0.0.1.
2184 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
2186 $thread_data{$thread}{'new_connection'} = 0;
2187 $stats{'reused-connections'}++;
2189 } elsif ($c =~ m/^Closing client socket \d+. .* Requests received: (\d+)\.$/) {
2191 # Closing client socket 12. Keep-alive: 1. Socket alive: 1. Data available: 0. \
2192 # Configuration file change detected: 0. Requests received: 14.
2194 $stats{'client-requests-on-connection'}{$1}++;
2195 $stats{'closed-client-connections'}++;
2199 sub gather_loglevel_header_stats($$) {
2201 my ($c, $thread) = @_;
2205 if ($c =~ m/^A HTTP\/1\.1 response without/ or
2206 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
2208 # A HTTP/1.1 response without Connection header implies keep-alive.
2209 # Keeping the server header 'Connection: keep-alive' around.
2210 $stats{'server-keep-alive'}++;
2219 'server-keep-alive' => 0,
2220 'reused-connections' => 0,
2221 'empty-responses' => 0,
2222 'empty-responses-on-new-connections' => 0,
2223 'empty-responses-on-reused-connections' => 0,
2224 'fast-redirections' => 0,
2226 'connection-failure' => 0,
2227 'connection-timeout' => 0,
2228 'reused-connections' => 0,
2229 'server-keep-alive' => 0,
2230 'closed-client-connections' => 0,
2231 'content-size-total' => 0,
2233 $stats{'client-requests-on-connection'}{1} = 0;
2236 sub get_percentage($$) {
2240 # If small is 0 the percentage is always 0%.
2241 # Make sure it works even if big is 0 as well.
2242 return "0.00%" if ($small eq 0);
2244 # Prevent division by zero.
2245 # XXX: Is this still supposed to be reachable?
2246 return "NaN" if ($big eq 0);
2248 return sprintf("%.2f%%", $small / $big * 100);
2255 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2256 my $client_requests_checksum = 0;
2259 if ($stats{requests_clf} && $stats{requests}
2260 && $stats{requests_clf} != $stats{requests}) {
2261 print "Inconsistent request counts: " . $stats{requests} . "/" . $stats{requests_clf} . "\n";
2264 # To get the total number of requests we can use either the number
2265 # of Common-Log-Format lines or the number of "Request:" messages.
2266 # We prefer the number of CLF lines if available because using
2267 # it works when analysing old log files from Privoxy versions before 3.0.29.
2268 # In Privoxy 3.0.28 and earlier "Request:" messages excluded
2269 # crunched messages.
2270 $requests_total = $stats{requests_clf} ? $stats{requests_clf} : $stats{requests};
2272 if ($requests_total eq 0) {
2273 print "No requests yet.\n";
2277 print "Client requests total: " . $requests_total . "\n";
2278 if ($stats{crunches}) {
2279 my $outgoing_requests = $requests_total - $stats{crunches};
2280 print "Crunches: " . $stats{crunches} . " (" .
2281 get_percentage($requests_total, $stats{crunches}) . ")\n";
2282 print "Blocks: " . $stats{'blocked'} . " (" .
2283 get_percentage($requests_total, $stats{'blocked'}) . ")\n";
2284 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2285 get_percentage($requests_total, $stats{'fast-redirections'}) . ")\n";
2286 print "Connection timeouts: " . $stats{'connection-timeout'} . " (" .
2287 get_percentage($requests_total, $stats{'connection-timeout'}) . ")\n";
2288 print "Connection failures: " . $stats{'connection-failure'} . " (" .
2289 get_percentage($requests_total, $stats{'connection-failure'}) . ")\n";
2290 print "Outgoing requests: " . $outgoing_requests . " (" .
2291 get_percentage($requests_total, $outgoing_requests) . ")\n";
2293 print "No crunches detected. Is 'debug 1024' enabled?\n";
2296 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2297 get_percentage($requests_total, $stats{'server-keep-alive'}) . ")\n";
2298 print "New outgoing connections: " . $new_connections . " (" .
2299 get_percentage($requests_total, $new_connections) . ")\n";
2300 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2301 get_percentage($requests_total, $stats{'reused-connections'}) .
2302 "; server offers accepted: " .
2303 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2304 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2305 get_percentage($requests_total, $stats{'empty-responses'}) . ")\n";
2306 print "Empty responses on new connections: "
2307 . $stats{'empty-responses-on-new-connections'} . " (" .
2308 get_percentage($requests_total, $stats{'empty-responses-on-new-connections'})
2310 print "Empty responses on reused connections: " .
2311 $stats{'empty-responses-on-reused-connections'} . " (" .
2312 get_percentage($requests_total, $stats{'empty-responses-on-reused-connections'}) .
2314 print "Client connections: " . $stats{'closed-client-connections'} . "\n";
2315 if ($stats{'content-size-total'}) {
2316 print "Bytes of content transfered to the client: " . $stats{'content-size-total'} . "\n";
2318 my $lines_printed = 0;
2319 print "Client requests per connection distribution:\n";
2320 foreach my $client_requests (sort {
2321 $stats{'client-requests-on-connection'}{$b} <=> $stats{'client-requests-on-connection'}{$a}}
2322 keys %{$stats{'client-requests-on-connection'}
2325 my $count = $stats{'client-requests-on-connection'}{$client_requests};
2326 $client_requests_checksum += $count * $client_requests;
2327 if ($cli_options{'show-complete-request-distribution'} or ($lines_printed < 10)) {
2328 printf "%8d: %d\n", $count, $client_requests;
2332 unless ($cli_options{'show-complete-request-distribution'}) {
2333 printf "Enable --show-complete-request-distribution to get less common numbers as well.\n";
2335 # Due to log rotation we may not have a complete picture for all the requests
2336 printf "Improperly accounted requests: ~%d\n", abs($requests_total - $client_requests_checksum);
2338 if (exists $stats{method}) {
2339 print "Method distribution:\n";
2340 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2341 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2344 print "Method distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2346 if (exists $stats{'http-version'}) {
2347 print "Client HTTP versions:\n";
2348 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2349 printf "%8d : %-8s\n", $stats{'http-version'}{$http_version}, $http_version;
2352 print "HTTP version distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2354 if (exists $stats{'status-code'}) {
2355 print "HTTP status codes:\n";
2356 foreach my $status_code (sort {$stats{'status-code'}{$b} <=> $stats{'status-code'}{$a}} keys %{$stats{'status-code'}}) {
2357 printf "%8d : %-8d\n", $stats{'status-code'}{$status_code}, $status_code;
2360 print "Status code distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2363 if ($cli_options{'url-statistics-threshold'} == 0) {
2364 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2366 print "Requested URLs:\n";
2367 foreach my $resource (sort {$stats{'resource'}{$b} <=> $stats{'resource'}{$a}} keys %{$stats{'resource'}}) {
2368 if ($stats{'resource'}{$resource} < $cli_options{'url-statistics-threshold'}) {
2369 print "Skipped statistics for URLs below the treshold.\n";
2372 printf "%d : %s\n", $stats{'resource'}{$resource}, $resource;
2376 if ($cli_options{'host-statistics-threshold'} == 0) {
2377 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2379 print "Requested Hosts:\n";
2380 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2381 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2382 print "Skipped statistics for Hosts below the treshold.\n";
2385 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2391 ################################################################################
2392 # Functions that actually print stuff
2393 ################################################################################
2395 sub print_clf_message() {
2397 our ($ip, $timestamp, $request_line, $status_code, $size);
2400 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2402 # Rebuild highlighted
2403 $output .= $h{'Number'} . $ip . $h{'Standard'};
2405 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2407 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2409 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2411 $output .= $h{'Number'} . $size . $h{'Standard'};
2412 $output .= $line_end;
2417 sub print_non_clf_message($) {
2419 my $content = shift;
2420 my $date_string = $keep_date_mode ? $req{$t}{'day'} . ' ' : '';
2421 my $msec_string = $no_msecs_mode ? '' : '.' . $req{$t}{'msecs'};
2422 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2424 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2428 . $time_colours[$time_colour_index % 2]
2429 . $req{$t}{'time-stamp'}
2431 . $h{Standard} . " "
2432 . $thread_colours{$t}
2436 . $h{$req{$t}{'log-level'}}
2437 . $req{$t}{'log-level'}
2444 sub shorten_thread_id($) {
2446 my $thread_id = shift;
2448 our %short_thread_ids;
2451 unless (defined $short_thread_ids{$thread_id}) {
2452 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2455 return $short_thread_ids{$thread_id}
2460 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2462 my $last_thread = 0;
2463 my $last_timestamp = 0;
2464 my $filters_that_did_nothing;
2467 $time_colour = paint_it('white');
2469 my %log_level_handlers = (
2470 'Re-Filter' => \&handle_loglevel_re_filter,
2471 'Header' => \&handle_loglevel_header,
2472 'Connect' => \&handle_loglevel_connect,
2473 'Redirect' => \&handle_loglevel_redirect,
2474 'Request' => \&handle_loglevel_request,
2475 'Crunch' => \&handle_loglevel_crunch,
2476 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2477 'Info' => \&handle_loglevel_info,
2478 'CGI' => \&handle_loglevel_cgi,
2479 'Force' => \&handle_loglevel_force,
2480 'Error' => \&handle_loglevel_error,
2481 'Fatal error' => \&handle_loglevel_ignore,
2482 'Writing' => \&handle_loglevel_ignore,
2483 'Received' => \&handle_loglevel_ignore,
2484 'Tagging' => \&handle_loglevel_tagging,
2485 'Actions' => \&handle_loglevel_ignore,
2486 'Unknown log level' => \&handle_loglevel_ignore,
2491 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2492 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2493 $req{$t}{'day'} = $day = $1;
2494 $req{$t}{'time-stamp'} = $time_stamp = $2;
2495 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2496 $req{$t}{'log-level'} = $log_level = $5;
2497 $req{$t}{'content'} = $content = $c = $6;
2498 $req{$t}{'log-message'} = $_;
2499 $no_special_header_highlighting = 0;
2501 if (defined($log_level_handlers{$log_level})) {
2503 $content = $log_level_handlers{$log_level}($content);
2507 die "No handler found for log level \"$log_level\"\n";
2510 # Highlight Truncations
2511 if (length($_) > 4000) {
2512 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2515 next unless $content;
2517 # Register threads to keep the colour constant
2518 if (!defined($thread_colours{$thread})) {
2519 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2520 $thread_colour_index++;
2523 # Switch timestamp colour if timestamps differ
2524 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2525 debug_message("Tick tack!") if DEBUG_TICKS;
2526 $time_colour = $time_colours[$time_colour_index % 2];
2527 $time_colour_index++;
2528 $last_msecs = $msecs;
2529 $last_timestamp = $time_stamp;
2532 $last_thread = $thread;
2534 print_non_clf_message($content);
2536 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2538 # LOG_LEVEL_CLF lines look like this
2539 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2540 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2541 # HTTP/1.1" 403 1730
2542 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2544 print_clf_message();
2548 # Some Privoxy log messages span more than one line,
2549 # usually to dump lots of content that doesn't need any syntax highlighting.
2550 # XXX: add mechanism to forward these lines to the right handler anyway.
2552 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2553 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2561 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2562 my $strict_checks = cli_option_is_set('strict-checks');
2563 my %log_level_handlers = (
2564 'Connect:' => \&gather_loglevel_connect_stats,
2565 'Crunch:' => \&gather_loglevel_crunch_stats,
2566 'Error:' => \&gather_loglevel_error_stats,
2567 'Header:' => \&gather_loglevel_header_stats,
2568 'Request:' => \&gather_loglevel_request_stats,
2570 my %ignored_log_levels = (
2571 'Actions:' => \&handle_loglevel_ignore,
2572 'CGI:' => \&handle_loglevel_ignore,
2573 'Fatal error:' => \&handle_loglevel_ignore,
2574 'Force:' => \&handle_loglevel_ignore,
2575 'Gif-Deanimate:' => \&handle_loglevel_ignore,
2576 'Info:' => \&handle_loglevel_ignore,
2577 'Re-Filter:' => \&handle_loglevel_ignore,
2578 'Received:' => \&handle_loglevel_ignore,
2579 'Redirect:' => \&handle_loglevel_ignore,
2580 'Unknown log level:' => \&handle_loglevel_ignore,
2581 'Writing:' => \&handle_loglevel_ignore,
2582 'Tagging:' => \&handle_loglevel_ignore,
2586 (undef, $time_stamp, $thread, $log_level, $content) = split(/ /, $_, 5);
2589 next if (not defined($log_level));
2591 if ($time_stamp eq "-") {
2593 gather_loglevel_clf_stats($content);
2595 } elsif (defined($log_level_handlers{$log_level})) {
2597 $content = $log_level_handlers{$log_level}($content, $thread);
2599 } elsif ($strict_checks and not defined($ignored_log_levels{$log_level})) {
2601 die "No handler found for: $_";
2609 sub unbreak_lines_only_loop() {
2610 my $log_messages_reached = 0;
2614 # Log level other than LOG_LEVEL_CLF?
2615 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2617 m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2618 $log_messages_reached = 1;
2623 $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
2628 print "\n" unless $log_messages_reached;
2633 sub VersionMessage {
2634 my $version_message;
2636 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2637 $version_message .= 'https://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2639 print $version_message;
2642 sub get_cli_options() {
2644 our %cli_options = (
2645 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2646 'title' => CLI_OPTION_TITLE,
2647 'keep-date' => CLI_OPTION_KEEP_DATE,
2648 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2649 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2650 'no-msecs' => CLI_OPTION_NO_MSECS,
2651 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2652 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2653 'statistics' => CLI_OPTION_STATISTICS,
2654 'strict-checks' => CLI_OPTION_STRICT_CHECKS,
2655 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2656 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2657 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2658 'show-complete-request-distribution' => CLI_OPTION_SHOW_COMPLETE_REQUEST_DISTRIBUTION,
2662 'html-output' => \$cli_options{'html-output'},
2663 'title' => \$cli_options{'title'},
2664 'keep-date' => \$cli_options{'keep-date'},
2665 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2666 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2667 'no-msecs' => \$cli_options{'no-msecs'},
2668 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2669 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2670 'statistics' => \$cli_options{'statistics'},
2671 'strict-checks' => \$cli_options{'strict-checks'},
2672 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2673 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
2674 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
2675 'show-complete-request-distribution' => \$cli_options{'show-complete-request-distribution'},
2676 'version' => sub { VersionMessage && exit(0) },
2680 $html_output_mode = cli_option_is_set('html-output');
2681 $no_msecs_mode = cli_option_is_set('no-msecs');
2682 $keep_date_mode = cli_option_is_set('keep-date');
2683 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2684 $line_end = get_line_end();
2695 Options and their default values if they have any:
2696 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2700 [--no-syntax-highlighting]
2701 [--shorten-thread-ids]
2702 [--show-ineffective-filters]
2703 [--show-complete-request-distribution]
2705 [--unbreak-lines-only]
2706 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2707 [--title $cli_options{'title'}]
2709 see "perldoc $0" for more information
2715 ################################################################################
2717 ################################################################################
2721 set_background(DEFAULT_BACKGROUND);
2722 prepare_our_stuff();
2726 # XXX: should explicitly reject incompatible argument combinations
2727 if (cli_option_is_set('unbreak-lines-only')) {
2728 unbreak_lines_only_loop();
2729 } elsif (cli_option_is_set('statistics')) {
2742 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2746 B<privoxy-log-parser> [B<--html-output>]
2747 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2748 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2749 [B<--url-statistics-threshold>] [B<--version>]
2753 B<privoxy-log-parser> reads Privoxy log messages and
2755 - syntax-highlights recognized lines,
2757 - reformats some of them for easier comprehension,
2759 - filters out less useful messages, and
2761 - (in some cases) calculates additional information,
2762 like the compression ratio or how a filter affected
2765 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2766 without getting confused by the resulting amount of output. For example for
2767 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2768 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2769 will hide the "filter foo caused 0 hits" message.
2773 [B<--host-statistics-threshold>] Only show the request count for a host
2774 if it's above or equal to the given threshold. If the threshold is 0, host
2775 statistics are disabled.
2777 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2778 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2779 This option is only intended to make embedding log excerpts in web pages easier.
2780 It does not escape any input!
2782 [B<--keep-date>] Don't remove the date when printing highlighted log messages.
2783 Useful when parsing multiple log files at once.
2785 [B<--no-msecs>] Don't expect milisecond resolution
2787 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2788 the filtered output is piped into less in which case the ANSI control
2789 codes don't work, or if the terminal itself doesn't support the control
2792 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2793 Note that the mapping from thread ids to shortened ids is created at run-time
2794 and thus varies with the input.
2796 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2797 that didn't modify the content.
2799 [B<--show-complete-request-distribution>] Show the complete client request
2800 distribution in the B<--statistics> output. Without this option only the
2801 ten most common numbers are shown.
2803 [B<--statistics>] Gather various statistics instead of syntax highlighting
2804 log messages. This is an experimental feature, if the results look wrong
2805 they very well might be. Also note that the results are pretty much guaranteed
2806 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2808 [B<--strict-checks>] When generating statistics, look more careful at the
2809 input data and abort if it is unexpected, even if it doesn't affect the
2810 results. Significantly slows the parsing down and is not expected to catch
2811 any problems that matter.
2812 When highlighting, print warnings in case of unknown messages which can't be
2813 properly highlighted.
2815 [B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
2816 interestingly configured mail client and thus are no longer recognized properly.
2817 Only fixes some breakage, but may be good enough or at least better than nothing.
2818 Doesn't do anything else, so you probably want to pipe the output into
2819 B<privoxy-log-parser> again.
2821 [B<--url-statistics-threshold>] Only show the request count for a resource
2822 if it's above or equal to the given threshold. If the threshold is 0, URL
2823 statistics are disabled.
2825 [B<--version>] Print version and exit.
2829 To monitor a log file:
2831 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2833 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2834 or if the log won't get rotated anyway. The log file location depends
2835 on your system (Doh!).
2837 To monitor Privoxy without having it write to a log file:
2839 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2841 Again, the config file location depends on your system. Output redirection
2842 depends on your shell, the above works with bourne shells.
2844 To read a processed Privoxy log file from top to bottom, letting the content
2845 scroll by slightly faster than you can read:
2847 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2849 This is probably only useful to fill screens in the background of haxor movies.
2853 Syntax highlighting with ANSI escape sequences will look strange
2854 if your background color isn't black.
2856 Some messages aren't recognized yet and will not be fully highlighted.
2858 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2859 using earlier Privoxy versions will probably result in an increased amount
2860 of unrecognized log lines.
2862 Privoxy's log files tend to be rather large. If you use HTML
2863 highlighting some browsers can't handle them, get confused and
2864 will eventually crash because of segmentation faults or unexpected
2865 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2870 Many settings can't be controlled through command line options yet.
2878 Fabian Keil <fk@fabiankeil.de>