3 ################################################################################
6 # A parser for Privoxy log messages. For incomplete documentation run
7 # perldoc privoxy-log-parser(.pl), for fancy screenshots see:
9 # http://www.fabiankeil.de/sourcecode/privoxy-log-parser/
11 # $Id: privoxy-log-parser.pl,v 1.69 2009/12/31 11:54:55 fabiankeil Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
27 # Copyright (c) 2007-2009 Fabian Keil <fk@fabiankeil.de>
29 # Permission to use, copy, modify, and distribute this software for any
30 # purpose with or without fee is hereby granted, provided that the above
31 # copyright notice and this permission notice appear in all copies.
33 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
34 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
35 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
36 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
37 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
38 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
39 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
40 ################################################################################
47 PRIVOXY_LOG_PARSER_VERSION => '0.5',
48 # Feel free to mess with these ...
49 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
50 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
51 HEADER_DEFAULT_COLOUR => 'yellow',
52 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
54 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
55 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
56 CLI_OPTION_NO_EMBEDDED_CSS => 0,
57 CLI_OPTION_NO_MSECS => 0,
58 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
59 CLI_OPTION_ERROR_LOG_FILE => '/var/log/privoxy-log.log',
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
62 CLI_OPTION_STATISTICS => 0,
64 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
66 SHOW_FILTER_READIN_IN => 0,
67 SUPPRESS_EMPTY_LINES => 1,
68 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
69 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
70 SUPPRESS_GIF_NOT_CHANGED => 1,
71 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
73 DEBUG_HEADER_REGISTERING => 0,
74 DEBUG_HEADER_HIGHLIGHTING => 0,
77 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
79 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
80 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
82 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
84 # You better leave these alone unless you know what you're doing.
85 COLOUR_RESET => "\033[0;0m",
89 # For performance reasons, these are global.
92 my %req; # request data from previous lines
97 my $thread_colour_index = 0;
98 my $header_colour_index = 0;
99 my $time_colour_index = 0;
101 my $no_special_header_highlighting;
104 my $header_highlight_regex = '';
106 my $html_output_mode;
107 my $no_msecs_mode; # XXX: should probably be removed
110 sub prepare_our_stuff () {
112 # Syntax Higlight hash
114 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
115 'light_gray', 'light_red', 'light_green', 'yellow',
116 'light_blue', 'pink', 'light_cyan', 'white'
123 Filter => 'purple', # XXX: Used?
124 'Re-Filter' => 'purple',
126 Request => 'light_cyan',
127 CGI => 'light_green',
129 Error => 'light_red',
131 'Fatal error' => 'light_red',
132 'Gif-Deanimate' => 'blue',
134 Writing => 'light_green',
135 # ----------------------
138 request_ => 'brown', # host+path but no protocol
139 'ip-address' => 'yellow',
142 Truncation => 'light_red',
144 Timestamp => 'brown',
145 Crunching => 'light_red',
146 crunched => 'light_red',
147 'Request-Line' => 'pink',
149 destination => 'yellow',
150 'http-version' => 'pink',
151 'crunch-pattern' => 'pink',
156 'program-name' => 'cyan',
159 warning => 'light_red',
160 debug => 'light_red',
164 'status-message' => 'light_cyan',
165 'status-code' => 'yellow',
166 'invalid-request' => 'light_red',
168 error => 'light_red',
169 'rewritten-URL' => 'light_red',
170 'pcrs-delimiter' => 'light_red',
171 'ignored' => 'light_red',
172 'action-bits-update' => 'light_red',
173 'configuration-line' => 'red',
174 'content-type' => 'yellow',
179 # Header colours need their own hash so the keys can be accessed properly
181 # Prefilled with headers that should not appear with default header colours
182 Cookie => 'light_red',
183 'Set-Cookie' => 'light_red',
184 Warning => 'light_red',
185 Default => HEADER_DEFAULT_COLOUR,
188 # Crunch reasons need their own hash as well
190 'Unsupported HTTP feature' => 'light_red',
191 Blocked => 'light_red',
192 Untrusted => 'light_red',
193 Redirected => 'green',
194 'CGI Call' => 'white',
195 'DNS failure' => 'red',
196 'Forwarding failed' => 'light_red',
197 'Connection failure' => 'light_red',
198 'Out of memory (may mask other reasons)' => 'light_red',
199 'No reason recorded' => 'light_red',
202 @time_colours = ('white', 'light_gray');
204 # Translate highlight strings into highlight code
205 prepare_highlight_hash(\%header_colours);
206 prepare_highlight_hash(\%reason_colours);
207 prepare_highlight_hash(\%h);
208 prepare_colour_array(\@all_colours);
209 prepare_colour_array(\@time_colours);
216 ###############################################################
217 # Takes a colour string and returns an ANSI escape sequence
218 # (unless --no-syntax-highlighting is used).
219 # XXX: The Rolling Stones reference has to go.
220 ###############################################################
222 my $colour = shift @_;
224 return "" if cli_option_is_set('no-syntax-highlighting');
266 my $bg_code = get_background();
268 our $default = default_colours();
270 if (defined($text{$colour})) {
271 $colour_code = ESCAPE;
272 $colour_code .= $text{$colour};
274 $colour_code .= $light{$colour} ? "1" : "2";
276 $colour_code .= $bg_code;
278 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
280 } elsif ($colour =~ /reset/) {
282 $colour_code = default_colours();
286 die "What's $colour supposed to mean?\n";
292 sub get_semantic_html_markup ($) {
293 ###############################################################
294 # Takes a string and returns a span element
295 ###############################################################
300 if ($type =~ /Standard/) {
304 $code = '<span title="' . $type . '" class="' . $type . '">';
310 sub cli_option_is_set ($) {
313 my $cli_option = shift;
315 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
317 return $cli_options{$cli_option};
320 sub get_html_title () {
323 return $cli_options{'title'};
327 sub init_css_colours() {
335 purple => "F06", # XXX: wrong
336 cyan => "F09", # XXX: wrong
341 light_green => "33F",
350 sub get_css_colour ($) {
355 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
357 return '#' . $css_colours{$colour};
360 sub get_css_line ($) {
365 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
366 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
367 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
368 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
369 $css_line .= '}' . "\n";
374 sub get_css_line_for_colour ($) {
379 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
380 $css_line .= 'color:' . get_css_colour($colour) . ';';
381 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
382 $css_line .= '}' . "\n";
387 # XXX: Wrong solution
388 sub get_missing_css_lines () {
392 $css_line .= '.' . 'default' . ' {';
393 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
394 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
395 $css_line .= '}' . "\n";
402 our %css_colours; #XXX: Wrong solution
406 $css .= '.privoxy-log {';
407 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
408 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
411 foreach my $key (keys %h_colours) {
413 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
414 $css .= get_css_line($key);
418 foreach my $colour (keys %css_colours) {
420 $css .= get_css_line_for_colour($colour);
424 $css .= get_missing_css_lines(); #XXX: Wrong solution
433 if (cli_option_is_set('html-output')) {
435 my $title = get_html_title();
437 $intro .= '<html><head>';
438 $intro .= '<title>' . $title . '</title>';
439 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
440 $intro .= '</head><body>';
441 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
451 if (cli_option_is_set('html-output')) {
453 $outro = '</p></body></html>';
459 sub get_line_end () {
460 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
463 sub get_colour_html_markup ($) {
464 ###############################################################
465 # Takes a colour string a span element. XXX: WHAT?
466 # XXX: This function shouldn't be necessary, the
467 # markup should always be semantically correct.
468 ###############################################################
473 if ($type =~ /Standard/) {
476 $code = '<span class="' . lc($type) . '">';
482 sub default_colours () {
485 return reset_colours();
488 sub show_colours () {
492 sub reset_colours () {
493 return ESCAPE . "0m";
496 sub set_background ($){
512 if (defined($backgrounds{$colour})) {
513 $bg_code = $backgrounds{$colour};
515 die "Invalid background colour: " . $colour;
519 sub get_background (){
523 sub prepare_highlight_hash ($) {
526 foreach my $key (keys %$ref) {
527 $$ref{$key} = $html_output_mode ?
528 get_semantic_html_markup($key) :
529 paint_it($$ref{$key});
533 sub prepare_colour_array ($) {
536 foreach my $i (0 ... @$ref - 1) {
537 $$ref[$i] = $html_output_mode ?
538 get_colour_html_markup($$ref[$i]) :
543 sub found_unknown_content ($) {
548 return if cli_option_is_set('accept-unknown-messages');
550 return if ($unknown =~ /\[too long, truncated\]$/);
552 $message = "found_unknown_content: Don't know how to highlight: ";
553 # Break line so the log file can later be parsed as Privoxy log file again
554 $message .= '"' . $unknown . '"' . " in:\n";
555 $message .= $req{$t}{'log-message'};
556 debug_message($message);
557 log_parse_error($req{$t}{'log-message'});
559 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
562 sub log_parse_error ($) {
566 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
567 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
568 print ERRORLOG $message;
573 sub debug_message (@) {
576 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
579 ################################################################################
580 # highlighter functions that aren't loglevel-specific
581 ################################################################################
585 # Get highlight marker
586 my $highlight = shift; # XXX: Stupid name;
590 if (defined($highlight)) {
592 $result = $h{$highlight};
596 $message = "h: Don't recognize highlighter $highlight.";
597 debug_message($message);
598 log_parser_error($message);
599 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
605 sub highlight_known_headers ($) {
609 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
611 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
613 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
614 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
620 sub highlight_matched_request_line ($$) {
622 my $result = shift; # XXX: Stupid name;
624 if ($result =~ m@(.*)($regex)(.*)@) {
625 $result = $1 . highlight_request_line($2) . $3
630 sub highlight_request_line ($) {
633 my ($method, $url, $http_version);
635 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
636 if ($rl =~ m/Invalid request/) {
638 $rl = h('invalid-request') . $rl . h('Standard');
640 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
642 # XXX: might not match in case of HTTP method fuzzing.
643 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
644 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
645 if ($rl =~ /http:\/\//) {
646 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
648 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
651 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
653 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
655 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
656 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
658 } elsif ($rl =~ m/^ $/) {
660 $rl = h('error') . "No request line specified!" . h('Standard');
664 debug_message ("Can't parse request line: $rl");
671 sub highlight_response_line ($) {
674 my ($http_version, $status_code, $status_message);
679 # TODO: Mark different status codes differently
681 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
682 ($http_version, $status_code, $status_message) = ($1, $2, $3);
684 debug_message ("Can't parse response line: $rl") and die 'Fix this';
687 # Rebuild highlighted
689 $rl .= h('http-version') . $http_version . h('Standard');
691 $rl .= h('status-code') . $status_code . h('Standard');
693 $rl .= h('status-message') . $status_message . h('Standard');
698 sub highlight_matched_url ($$) {
700 my $result = shift; # XXX: Stupid name;
703 #print "Got $result, regex ($regex)\n";
705 if ($result =~ m@(.*?)($regex)(.*)@) {
706 $result = $1 . highlight_url($2) . $3;
707 #print "Now the result is $result\n";
713 sub highlight_matched_host ($$) {
715 my ($result, $regex) = @_; # XXX: result ist stupid name;
717 if ($result =~ m@(.*?)($regex)(.*)@) {
718 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
724 sub highlight_matched_pattern ($$$) {
726 my $result = shift; # XXX: Stupid name;
730 die "Unknown key $key" unless defined $h{$key};
732 if ($result =~ m@(.*?)($regex)(.*)@) {
733 $result = $1 . h($key) . $2 . h('Standard') . $3;
739 sub highlight_matched_path ($$) {
741 my $result = shift; # XXX: Stupid name;
744 if ($result =~ m@(.*?)($regex)(.*)@) {
745 $result = $1 . h('path') . $2 . h('Standard') . $3;
751 sub highlight_url ($) {
755 if ($html_output_mode) {
757 $url = '<a href="' . $url . '">' . $url . '</a>';
761 $url = h('URL') . $url . h('Standard');
768 sub update_header_highlight_regex ($) {
771 my $headers = join ('|', keys %header_colours);
773 $header_highlight_regex = qr/$headers/;
774 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
777 ################################################################################
778 # loglevel-specific highlighter functions
779 ################################################################################
781 sub handle_loglevel_header ($) {
785 if ($c =~ /^scan:/) {
787 if ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
789 # Client request line
790 # Save for statistics (XXX: Not implemented yet)
791 $req{$t}{'method'} = $2;
792 $req{$t}{'destination'} = $3;
793 $req{$t}{'http-version'} = $4;
795 $c = highlight_request_line($1);
797 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
799 # Server response line
800 $req{$t}{'response_line'} = $2;
801 $req{$t}{'status_code'} = $3;
802 $req{$t}{'status_message'} = $4;
803 $c = $1 . highlight_response_line($req{$t}{'response_line'});
805 } elsif ($c =~ m/^scan: ((?>[^:]+)):/) {
807 # Register new headers
808 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
810 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
811 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
813 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
814 $header_colours{$header} = $header_colours{'Default'};
816 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
817 $header_colour_index++;
819 update_header_highlight_regex($header);
824 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
826 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
827 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
828 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
830 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
832 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
833 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+');
834 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
836 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
838 # Text mode enabled by force. Take cover!
839 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
841 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
843 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
844 $c = $1 . highlight_request_line($2);
846 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
848 # Adjusted Content-Length to 2132
849 # Adjust Content-Length to 33533
850 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
851 $c = highlight_known_headers($c);
853 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
855 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
856 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
858 } elsif ($c =~ m/^Couldn\'t parse:/) {
860 # XXX: These should probable be logged with LOG_LEVEL_ERROR
861 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
862 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
863 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
865 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
866 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
868 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
869 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
870 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
872 # XXX: Save tag and tagger
874 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
875 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
876 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
877 $no_special_header_highlighting = 1;
879 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
881 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
882 # XXX: Save tag and tagger
884 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
885 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
887 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
888 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
889 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
890 or $c =~ m/^(addh-unique|Referer forged to)/
891 or $c =~ m/^Downgraded answer to HTTP\/1.0/
892 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
893 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
894 or $c =~ m/^Referer crunched!/
895 or $c =~ m/^crunched x-forwarded-for!/
896 or $c =~ m/^crunched From!/
897 or $c =~ m/^ modified$/
898 or $c =~ m/^Content filtering is enabled. Crunching:/
899 or $c =~ m/^force-text-mode overruled the client/
900 or $c =~ m/^Server time in the future\./
901 or $c =~ m/^content-disposition header crunched and replaced with:/i
902 or $c =~ m/^Reducing white space in /
903 or $c =~ m/^Ignoring single quote in /
904 or $c =~ m/^Converting tab to space in /
905 or $c =~ m/A HTTP\/1\.1 response without/
906 or $c =~ m/Disabled filter mode on behalf of the client/
907 or $c =~ m/Keeping the (?:server|client) header /
908 or $c =~ m/Content modified with no Content-Length header set/
909 or $c =~ m/^Appended client IP address to/
910 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
911 or $c =~ m/^keep-alive support is disabled/
914 # XXX: Some of these may need highlighting
916 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
917 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
918 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
919 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
920 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
921 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
922 # Text mode is already enabled
923 # Denied request with NULL byte(s) turned into line break(s)
924 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
925 # addh-unique: Host: people.freebsd.org
926 # Suppressed offer to compress content
927 # Crunched incoming cookie -- yum!
928 # Accepted the client's request to fetch without filtering.
929 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
930 # addh-unique: Host: subkeys.pgp.net:11371
931 # Referer forged to: Referer: http://10.0.0.1/
932 # Downgraded answer to HTTP/1.0
933 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
934 # Referer overwritten with: Referer: pille-palle
935 # Referer replaced with: Referer: pille-palle
936 # crunched x-forwarded-for!
938 # modified # XXX: pretty stupid log message
939 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
940 # force-text-mode overruled the client's request to fetch without filtering!
941 # Server time in the future.
942 # content-disposition header crunched and replaced with: content-disposition: filename=baz
943 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
944 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
945 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
946 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
948 # A HTTP/1.1 response without Connection header implies keep-alive.
949 # Disabled filter mode on behalf of the client.
950 # Keeping the server header 'Connection: keep-alive' around.
951 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
952 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
953 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
954 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
955 # Removing 'Connection: close' to imply keep-alive.
956 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
958 } elsif ($c =~ m/^scanning headers for:/) {
960 return '' unless SHOW_SCAN_INTRO;
962 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
963 # crunched User-Agent!
964 # Crunching: Content-Encoding: gzip
966 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
968 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
970 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
972 $c = h('warning') . $c . h('Standard');
974 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
976 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
977 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
979 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
980 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
982 } elsif ($c =~ m/^Removing empty header/) {
984 # Removing empty header
987 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
989 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
990 # Enable force-text-mode if you know what you're doing.
991 # XXX: Could highlight more here.
992 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
994 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
996 # Server keep-alive timeout is 5. Sticking with 10.
997 # Client keep-alive timeout is 20. Sticking with 10.
999 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1000 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1002 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1004 # Reducing keep-alive timeout from 60 to 10.
1006 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1007 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1011 found_unknown_content($c);
1015 unless ($c =~ m/^Transforming/) {
1016 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1022 sub handle_loglevel_re_filter ($) {
1024 my $content = shift;
1028 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1030 # XXX: only the second version gets highlighted properly.
1031 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1032 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1033 $req{$t}{'content_source'} = $1;
1034 $req{$t}{'content_size'} = $2;
1035 $req{$t}{'content_filter'} = $3;
1036 $req{$t}{'content_hits'} = $4;
1037 $req{$t}{'new_content_size'} = $5;
1038 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1039 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1040 if ($req{$t}{'content_hits'} == 0 and
1041 not (cli_option_is_set('show-ineffective-filters')
1042 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1046 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1047 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1048 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1050 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1051 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1054 $c .= "(" . $h{'Number'};
1055 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1056 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1059 } elsif ($c =~ /\.{3}$/
1060 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1062 # Used by Privoxy 3.0.5 and 3.0.6:
1064 # Used by Privoxy 3.0.7:
1065 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1067 $req{$t}{'filtered_header'} = $1;
1068 $req{$t}{'old_header_size'} = $2;
1069 $req{$t}{'header_filter_name'} = $3;
1071 unless (cli_option_is_set('show-ineffective-filters') or
1072 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1075 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1076 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1078 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1080 # ...produced 0 hits (new size 23).
1081 #... produced 1 hits (new size 54).
1083 $req{$t}{'header_filter_hits'} = $1;
1084 $req{$t}{'new_header_size'} = $2;
1086 unless (cli_option_is_set('show-ineffective-filters') or
1087 (defined($req{$t}{'header_filter_name'}) and
1088 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1090 if ($req{$t}{'header_filter_hits'} == 0 and
1091 not (defined($req{$t}{'header_filter_name'}) and
1092 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1095 # Reformat including information from the intro
1096 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1098 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1099 # $c .= $req{$t}{'filtered_header'} . " ";
1100 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1101 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1103 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1105 $c .= "changing size from ";
1106 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1108 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1113 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1117 # Highlight from last line (XXX: What?)
1118 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1119 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1124 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1125 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1130 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1132 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1133 # Tagger variable-test has empty joblist. Nothing to do.
1135 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1137 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1139 $req{$t}{'chunked-size'} = $1;
1140 $req{$t}{'dechunked-size'} = $2;
1141 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1143 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1145 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1146 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1148 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1150 # Decompression successful. Old size: 670, new size: 1166.
1152 $req{$t}{'size-compressed'} = $1;
1153 $req{$t}{'size-decompressed'} = $2;
1154 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1156 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1157 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1159 # XXX: Create sub get_percentage()
1160 if ($req{$t}{'size-decompressed'}) {
1161 $req{$t}{'decompression-gain-percent'} =
1162 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1164 $content .= " (saved: ";
1165 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1167 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1171 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1173 # Need to de-chunk first
1174 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1176 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1178 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1180 # Adding re_filter job ...
1181 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1182 # to filter client-ip-address succeeded.
1184 } elsif ($c =~ m/^Reading in filter/) {
1186 return '' unless SHOW_FILTER_READIN_IN;
1190 found_unknown_content($content);
1197 sub handle_loglevel_redirect ($) {
1201 if ($c =~ m/^Decoding "([^""]*)"/) {
1203 $req{$t}{'original-destination'} = $1;
1204 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1207 } elsif ($c =~ m/^Checking/) {
1209 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1210 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1212 # TODO: Change colour if really url-decoded
1213 $req{$t}{'decoded-original-destination'} = $1;
1214 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1217 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1219 # pcrs command "s@&from=rss@@" changed \
1220 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1221 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1222 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1223 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1224 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1225 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1226 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1227 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1229 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1231 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1232 # change "http://www.example.org/".
1233 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1234 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1236 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1238 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1239 # XXX: Use URL highlighter
1241 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1243 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1244 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1245 # is already properly formatted.
1246 # XXX: assume the same?
1247 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1251 found_unknown_content($c);
1258 sub handle_loglevel_gif_deanimate ($) {
1260 my $content = shift;
1262 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1264 my $bytes_from = $1;
1266 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1267 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1268 # XXX: Do we need g in case of ($1 == $2)?
1269 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1271 } elsif ($content =~ m/GIF (not) changed/) {
1273 # Gif-Deanimate: GIF not changed.
1274 return '' if SUPPRESS_GIF_NOT_CHANGED;
1275 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1277 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1279 # failed! (gif parsing)
1280 # XXX: Replace this error message with something less stupid
1281 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1283 } elsif ($content =~ m/^Need to de-chunk first/) {
1285 # Need to de-chunk first
1286 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1288 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1290 # No GIF header found (XXX: Did I ever commit this?)
1291 # failed while parsing 195 134747048 (XXX: never commited)
1293 # Ignore these for now
1297 found_unknown_content($content);
1304 sub handle_loglevel_request ($) {
1306 my $content = shift;
1308 if ($content =~ m/crunch! /) {
1310 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1312 # Highlight crunch reasons
1313 foreach my $reason (keys %reason_colours) {
1314 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1316 # Highlight request URL domain and ditch 'crunch!'
1317 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1318 $content =~ s@ crunch!@@;
1320 } elsif ($content =~ m/\[too long, truncated\]$/) {
1322 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1323 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1325 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1327 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1328 $content = h('request_') . $content . h('Standard');
1332 found_unknown_content($content);
1339 sub handle_loglevel_crunch ($) {
1341 my $content = shift;
1343 # Highlight crunch reason
1344 foreach my $reason (keys %reason_colours) {
1345 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1348 if ($content =~ m/\[too long, truncated\]$/) {
1350 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1351 # [...]&filter... [too long, truncated]
1352 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1356 # Blocked: http://ads.example.org/
1357 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1363 sub handle_loglevel_connect ($) {
1367 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1369 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1371 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1372 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1374 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1376 # connect to: www.example.org.noconnect failed: Operation not permitted
1378 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1380 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1382 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1384 # Connect: to www.nzherald.co.nz successful
1385 # Connect: to archiv.radiotux.de
1387 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1388 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1390 } elsif ($c =~ m/^to ([^\s]*)$/) {
1392 # Connect: to lists.sourceforge.net:443
1394 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1396 } elsif ($c =~ m/^accepted connection from .*/ or
1399 # accepted connection from 10.0.0.1
1400 # Privoxy 3.0.6 and earlier just say:
1402 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1403 $c = highlight_matched_host($c, '(?<=connection from ).*');
1405 } elsif ($c =~ m/^write header to: .* failed:/) {
1407 # write header to: 10.0.0.1 failed: Broken pipe
1409 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1410 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1412 } elsif ($c =~ m/^write header to client failed:/) {
1414 # write header to client failed: Broken pipe
1416 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1418 } elsif ($c =~ m/^socks4_connect:/) {
1420 # socks4_connect: SOCKS request rejected or failed.
1421 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1423 } elsif ($c =~ m/^Listening for new connections/ or
1424 $c =~ m/^accept connection/) {
1426 # Privoxy versions above 3.0.6 say:
1427 # Listening for new connections ...
1428 # earlier versions say:
1429 # accept connection ...
1432 } elsif ($c =~ m/^accept failed:/) {
1434 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1436 } elsif ($c =~ m/^Overriding forwarding settings/) {
1438 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1439 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1441 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1443 # Denying suspicious CONNECT request from 10.0.0.1
1444 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1446 } elsif ($c =~ m/^socks5_connect:/) {
1448 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1450 } elsif ($c =~ m/^Created new connection to/) {
1452 # Created new connection to www.privoxy.org:80 on socket 11.
1453 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1454 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1456 } elsif ($c =~ m/^Found reusable socket/) {
1458 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1460 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1461 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1462 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1463 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1464 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1465 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1466 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1467 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1469 } elsif ($c =~ m/^Marking open socket/) {
1471 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1472 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1473 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1474 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1476 } elsif ($c =~ m/^No reusable/) {
1478 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1479 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1481 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1483 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1484 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1486 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1487 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1488 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 } elsif ($c =~ m/^Socket/) {
1492 # Socket 16 already forgotten or never remembered.
1493 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1495 } elsif ($c =~ m/^The connection to/) {
1497 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1499 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1500 # Timeout is: 1. Assumed latency: 4.
1501 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1502 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1503 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1504 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1505 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1506 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1508 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1510 # Stopped waiting for the request line. Timeout: 121.
1511 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1513 } elsif ($c =~ m/^Waiting for \d/) {
1515 # Waiting for 1 connections to timeout.
1516 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1518 } elsif ($c =~ m/^Initialized/) {
1520 # Initialized 20 socket slots.
1521 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1523 } elsif ($c =~ m/^Done reading from server/) {
1525 # Done reading from server. Expected content length: 24892. \
1526 # Actual content length: 24892. Most recently received: 4412.
1528 # Done reading from server. Expected content length: 24892. \
1529 # Actual content length: 24892. Bytes most recently read: 4412.
1530 # Done reading from server. Content length: 6018 as expected. \
1531 # Bytes most recently read: 294.
1532 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1533 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1534 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1536 } elsif ($c =~ m/^Continuing buffering headers/) {
1538 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1539 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1540 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1541 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1543 # Continuing buffering headers. Bytes most recently read: %d.
1544 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1546 } elsif ($c =~ m/^Received \d+ bytes while/) {
1548 # Received 206 bytes while expecting 12103.
1549 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1550 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 } elsif ($c =~ m/^Connection from/) {
1554 # Connection from 81.163.28.218 dropped due to ACL
1555 $c =~ s@(?<=^Connection from )((?:\d+\.?){4})@$h{'Number'}$1$h{'Standard'}@;
1557 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1558 $c =~ m/^No additional client request/) {
1560 # Reusing server socket 4. Opened for 10.0.0.1.
1561 # Closing server socket 2. Opened for 10.0.0.1.
1562 # No additional client request received in time. \
1563 # Closing server socket 4, initially opened for 10.0.0.1.
1564 # No additional client request received in time on socket 29.
1566 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1569 } elsif ($c =~ m/^Connected to /) {
1571 # Connected to tor-jail[10.0.0.2]:9050.
1573 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1574 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1575 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1577 } elsif ($c =~ m/^Could not connect to /) {
1579 # Could not connect to [10.0.0.1]:80.
1581 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1582 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1584 } elsif ($c =~ m/^Waiting for the next client request/ or
1585 $c =~ m/^The connection on server socket/ or
1586 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1588 # Waiting for the next client request on socket 3. Keeping the server \
1589 # socket 12 to a.fsdn.com open.
1590 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1591 # Client request arrived in time or the client closed the connection on socket 12.
1593 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1594 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1595 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1597 } elsif ($c =~ m/^Marking the server socket/) {
1599 # Marking the server socket 7 tainted.
1601 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1603 } elsif ($c =~ m/^Reduced expected bytes to /) {
1605 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1606 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1609 } elsif ($c =~ m/^The client closed socket /) {
1611 # The client closed socket 2 while the server socket 4 is still open.
1612 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1613 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1615 } elsif ($c =~ m/^Expected client content length set /) {
1617 # Expected client content length set to 667325411 after reading 4999 bytes.
1618 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1621 } elsif ($c =~ m/^Waiting for up to /) {
1623 # Waiting for up to 4999 bytes from the client.
1624 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1626 } elsif ($c =~ m/^Looks like we rea/ or
1627 $c =~ m/^Unsetting keep-alive flag/ or
1628 $c =~ m/^No connections to wait/ or
1629 $c =~ m/^Complete client request received/ or
1630 $c =~ m/^Possible pipeline attempt detected./ or
1631 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1632 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1633 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1634 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1635 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1636 $c =~ m/^Done reading from the client\.$/) {
1638 # Looks like we reached the end of the last chunk. We better stop reading.
1639 # Looks like we read the end of the last chunk together with the server \
1640 # headers. We better stop reading.
1641 # Unsetting keep-alive flag.
1642 # No connections to wait for left.
1643 # Client request arrived in time or the client closed the connection.
1644 # Complete client request received
1645 # Possible pipeline attempt detected. The connection will not be \
1646 # kept alive and we will only serve the first request.
1647 # POST request detected. The connection will not be kept alive.
1648 # The server still wants to talk, but the client hung up on us.
1649 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1650 # There might be a request body. The connection will not be kept alive.
1651 # Stopping to watch the client socket. There's already another request waiting.
1652 # Done reading from the client\.
1656 found_unknown_content($c);
1664 sub handle_loglevel_info ($) {
1668 if ($c =~ m/^Rewrite detected:/) {
1670 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1671 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1673 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1674 $c =~ m/^Compressed content detected/ or
1677 # Decompressing deflated iob: 117
1678 # Decompression didn't result in any content.
1679 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1680 # with zlib support or enable the prevent-compression action.
1681 # Tagger 'complete-url' created empty tag. Ignored.
1685 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1687 # loading configuration file '/usr/local/etc/privoxy/config':
1688 # Reloading configuration file '/usr/local/etc/privoxy/config'
1689 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1691 } elsif ($c =~ m/^exiting by signal/) {
1693 # exiting by signal 15 .. bye
1694 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1696 } elsif ($c =~ m/^Privoxy version/) {
1698 # Privoxy version 3.0.7
1699 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1701 } elsif ($c =~ m/^Program name: /) {
1703 # Program name: /usr/local/sbin/privoxy
1704 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1706 } elsif ($c =~ m/^Listening on port /) {
1708 # Listening on port 8118 on IP address 10.0.0.1
1709 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1710 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1712 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1714 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1715 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1717 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1719 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1720 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1721 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1723 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1724 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1725 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1726 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1728 } elsif ($c =~ m/^Status code/) {
1730 # Status code 304 implies no body.
1731 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1733 } elsif ($c =~ m/^Method/) {
1735 # Method HEAD implies no body.
1736 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1738 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1740 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1741 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1742 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1744 } elsif ($c =~ m/^No logfile configured/ or
1745 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1746 $c =~ m/^Invalid \"chunked\" transfer/ or
1747 $c =~ m/^Support for/ or
1748 $c =~ m/^Flushing header and buffers/
1751 # No logfile configured. Please enable it before reporting any problems.
1752 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1753 # response or even no response at all.
1754 # No logfile configured. Logging disabled.
1755 # Invalid "chunked" transfer encoding detected and ignored.
1756 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1757 # known not to work properly in some situations.
1758 # Flushing header and buffers. Stepping back from filtering.
1762 found_unknown_content($c);
1769 sub handle_loglevel_cgi ($) {
1773 if ($c =~ m/^Granting access to/) {
1775 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1777 } elsif ($c =~ m/^Substituting: s(.)/) {
1779 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1780 # XXX: prone to span several lines
1783 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1784 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1785 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1791 sub handle_loglevel_force ($) {
1795 if ($c =~ m/^Ignored force prefix in request:/) {
1797 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1798 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1799 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1801 } elsif ($c =~ m/^Enforcing request:/) {
1803 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1804 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1808 found_unknown_content($c);
1815 sub handle_loglevel_error ($) {
1819 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1821 # Empty server or forwarder response received on socket 4.
1822 # Empty server or forwarder response received on socket 3. \
1823 # Closing client socket 15 without sending data.
1824 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1825 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1827 # XXX: There are probably more messages that deserve highlighting.
1833 sub handle_loglevel_ignore ($) {
1837 sub gather_loglevel_request_stats ($$) {
1845 sub gather_loglevel_crunch_stats ($$) {
1855 sub gather_loglevel_error_stats ($$) {
1862 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1864 # Empty server or forwarder response received on socket 4.
1865 $stats{'empty-responses'}++;
1866 if ($thread_data{$thread}{'new_connection'}) {
1867 $stats{'empty-responses-on-new-connections'}++;
1869 $stats{'empty-responses-on-reused-connections'}++;
1874 sub gather_loglevel_connect_stats ($$) {
1881 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1883 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1884 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1886 } elsif ($c =~ m/^to ([^\s]*)$/) {
1888 # Connect: to lists.sourceforge.net:443
1890 $thread_data{$thread}{'forwarder'} = 'direct connection';
1892 } elsif ($c =~ m/^Created new connection to/) {
1894 # Created new connection to www.privoxy.org:80 on socket 11.
1896 $thread_data{$thread}{'new_connection'} = 1;
1898 } elsif ($c =~ m/^Reusing server socket \d./ or
1899 $c =~ m/^Found reusable socket/) {
1901 # Reusing server socket 4. Opened for 10.0.0.1.
1902 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1904 $thread_data{$thread}{'new_connection'} = 0;
1905 $stats{'reused-connections'}++;
1909 sub gather_loglevel_header_stats ($) {
1915 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1916 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1918 # A HTTP/1.1 response without Connection header implies keep-alive.
1919 # Keeping the server header 'Connection: keep-alive' around.
1920 $stats{'server-keep-alive'}++;
1928 'server-keep-alive' => 0,
1929 'reused-connections' => 0,
1930 'empty-responses' => 0,
1931 'empty-responses-on-new-connections' => 0,
1932 'empty-responses-on-reused-connections' => 0,
1936 sub get_percentage ($$) {
1939 return "NaN" if ($big eq 0);
1940 return sprintf("%.2f%%", $small / $big * 100);
1943 sub print_stats () {
1946 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1947 my $outgoing_requests = $stats{requests} - $stats{crunches};
1949 if ($stats{requests} eq 0) {
1950 print "No requests yet.\n";
1954 print "Client requests total: " . $stats{requests} . "\n";
1955 print "Crunches: " . $stats{crunches} . " (" .
1956 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
1957 print "Outgoing requests: " . $outgoing_requests . " (" .
1958 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
1959 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
1960 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
1961 print "New outgoing connections: " . $new_connections . " (" .
1962 get_percentage($stats{requests}, $new_connections) . ")\n";
1963 print "Reused connections: " . $stats{'reused-connections'} . " (" .
1964 get_percentage($stats{requests}, $stats{'reused-connections'}) . ")\n";
1965 print "Empty responses: " . $stats{'empty-responses'} . " (" .
1966 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
1967 print "Empty responses on new connections: "
1968 . $stats{'empty-responses-on-new-connections'} . " (" .
1969 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
1971 print "Empty responses on reused connections: " .
1972 $stats{'empty-responses-on-reused-connections'} . " (" .
1973 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
1978 ################################################################################
1979 # Functions that actually print stuff
1980 ################################################################################
1982 sub print_clf_message () {
1984 our ($ip, $timestamp, $request_line, $status_code, $size);
1987 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1989 # Rebuild highlighted
1990 $output .= $h{'Number'} . $ip . $h{'Standard'};
1992 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
1994 $output .= "\"" . highlight_request_line("$request_line") . "\"";
1996 $output .= $h{'Status'} . $status_code . $h{'Standard'};
1998 $output .= $h{'Number'} . $size . $h{'Standard'};
1999 $output .= $line_end;
2004 sub print_non_clf_message ($) {
2006 my $content = shift;
2007 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2008 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2010 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2013 . $time_colours[$time_colour_index % 2]
2014 . $req{$t}{'time-stamp'}
2016 . $h{Standard} . " "
2017 . $thread_colours{$t}
2021 . $h{$req{$t}{'log-level'}}
2022 . $req{$t}{'log-level'}
2031 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2033 my $last_thread = 0;
2034 my $last_timestamp = 0;
2035 my $filters_that_did_nothing;
2038 $time_colour = paint_it('white');
2040 my %log_level_handlers = (
2041 'Re-Filter' => \&handle_loglevel_re_filter,
2042 'Header' => \&handle_loglevel_header,
2043 'Connect' => \&handle_loglevel_connect,
2044 'Redirect' => \&handle_loglevel_redirect,
2045 'Request' => \&handle_loglevel_request,
2046 'Crunch' => \&handle_loglevel_crunch,
2047 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2048 'Info' => \&handle_loglevel_info,
2049 'CGI' => \&handle_loglevel_cgi,
2050 'Force' => \&handle_loglevel_force,
2051 'Error' => \&handle_loglevel_error,
2052 'Fatal error' => \&handle_loglevel_ignore,
2053 'Writing' => \&handle_loglevel_ignore,
2054 'Unknown log level' => \&handle_loglevel_ignore,
2059 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2061 $req{$t}{'day'} = $day = $1;
2062 $req{$t}{'time-stamp'} = $time_stamp = $2;
2063 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2064 $req{$t}{'log-level'} = $log_level = $5;
2065 $req{$t}{'content'} = $content = $c = $6;
2066 $req{$t}{'log-message'} = $_;
2067 $no_special_header_highlighting = 0;
2069 if (defined($log_level_handlers{$log_level})) {
2071 $content = $log_level_handlers{$log_level}($content);
2075 die "No handler found for log level \"$log_level\"\n";
2078 # Highlight Truncations
2079 if (length($_) > 4000) {
2080 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2083 next unless $content;
2085 # Register threads to keep the colour constant
2086 if (!defined($thread_colours{$thread})) {
2087 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2088 $thread_colour_index++;
2091 # Switch timestamp colour if timestamps differ
2092 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2093 debug_message("Tick tack!") if DEBUG_TICKS;
2094 $time_colour = $time_colours[$time_colour_index % 2];
2095 $time_colour_index++;
2096 $last_msecs = $msecs;
2097 $last_timestamp = $time_stamp;
2100 $last_thread = $thread;
2102 print_non_clf_message($content);
2104 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2106 # LOG_LEVEL_CLF lines look like this
2107 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2108 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2109 # HTTP/1.1" 403 1730
2110 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2112 print_clf_message();
2116 # Some Privoxy log messages span more than one line,
2117 # usually to dump lots of content that doesn't need any syntax highlighting.
2118 # XXX: add mechanism to forward these lines to the right handler anyway.
2120 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2121 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2129 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2130 my %log_level_handlers = (
2131 'Re-Filter' => \&handle_loglevel_ignore,
2132 'Header' => \&gather_loglevel_header_stats,
2133 'Connect' => \&gather_loglevel_connect_stats,
2134 'Redirect' => \&handle_loglevel_ignore,
2135 'Request' => \&gather_loglevel_request_stats,
2136 'Crunch' => \&gather_loglevel_crunch_stats,
2137 'Gif-Deanimate' => \&handle_loglevel_ignore,
2138 'Info' => \&handle_loglevel_ignore,
2139 'CGI' => \&handle_loglevel_ignore,
2140 'Force' => \&handle_loglevel_ignore,
2141 'Error' => \&gather_loglevel_error_stats,
2142 'Fatal error' => \&handle_loglevel_ignore,
2143 'Writing' => \&handle_loglevel_ignore,
2144 'Unknown log level' => \&handle_loglevel_ignore
2148 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2151 $msecs = $3 ? $3 : 0;
2156 if (defined($log_level_handlers{$log_level})) {
2158 $content = $log_level_handlers{$log_level}($content, $thread);
2162 die "No handler found for log level \"$log_level\"\n";
2172 sub VersionMessage {
2173 my $version_message;
2175 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2176 $version_message .= 'Copyright (C) 2007-2009 Fabian Keil <fk@fabiankeil.de>' . "\n";
2177 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2179 print $version_message;
2182 sub get_cli_options () {
2184 our %cli_options = (
2185 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2186 'title' => CLI_OPTION_TITLE,
2187 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2188 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2189 'no-msecs' => CLI_OPTION_NO_MSECS,
2190 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2191 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2192 'statistics' => CLI_OPTION_STATISTICS,
2196 'html-output' => \$cli_options{'html-output'},
2197 'title' => \$cli_options{'title'},
2198 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2199 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2200 'no-msecs' => \$cli_options{'no-msecs'},
2201 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2202 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2203 'statistics' => \$cli_options{'statistics'},
2204 'version' => sub { VersionMessage && exit(0) },
2208 $html_output_mode = cli_option_is_set('html-output');
2209 $no_msecs_mode = cli_option_is_set('no-msecs');
2210 $line_end = get_line_end();
2221 Options and their default values if they have any:
2222 [--accept-unknown-messages]
2226 [--no-syntax-highlighting]
2227 [--show-ineffective-filters]
2229 [--title $cli_options{'title'}]
2231 see "perldoc $0" for more information
2237 ################################################################################
2239 ################################################################################
2243 set_background(DEFAULT_BACKGROUND);
2244 prepare_our_stuff();
2248 if (cli_option_is_set('statistics')) {
2261 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2265 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2266 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
2271 B<privoxy-log-parser> reads Privoxy log messages and
2273 - syntax-highlights recognized lines,
2275 - reformats some of them for easier comprehension,
2277 - filters out less useful messages, and
2279 - (in some cases) calculates additional information,
2280 like the compression ratio or how a filter affected
2283 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2284 without getting confused by the resulting amount of output. For example for
2285 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2286 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2287 will hide the "filter foo caused 0 hits" message.
2291 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2292 just don't highlight them.
2294 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2295 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2296 This option is only intended to make embedding log excerpts in web pages easier.
2297 It does not escape any input!
2299 [B<--no-msecs>] Don't expect milisecond resolution
2301 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2302 the filtered output is piped into less in which case the ANSI control
2303 codes don't work, or if the terminal itself doesn't support the control
2306 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2307 that didn't modify the content.
2309 [B<--statistics>] Gather various statistics instead of syntax highlighting
2310 log messages. This is an experimental feature, if the results look wrong
2311 they very well might be. Also note that the results a pretty much guaranteed
2312 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2314 [B<--version>] Print version and exit.
2318 To monitor a log file:
2320 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2322 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2323 or if the log won't get rotated anyway. The log file location depends
2324 on your system (Doh!).
2326 To monitor Privoxy without having it write to a log file:
2328 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2330 Again, the config file location depends on your system. Output redirection
2331 depends on your shell, the above works with bourne shells.
2333 To read a processed Privoxy log file from top to bottom, letting the content
2334 scroll by slightly faster than you can read:
2336 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2338 This is probably only useful to fill screens in the background of haxor movies.
2342 Syntax highlighting with ANSI escape sequences will look strange
2343 if your background color isn't black.
2345 Some messages aren't recognized yet and will not be fully highlighted.
2347 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2348 using earlier Privoxy versions will probably result in an increased amount
2349 of unrecognized log lines.
2351 Privoxy's log files tend to be rather large. If you use HTML
2352 highlighting some browsers can't handle them, get confused and
2353 will eventually crash because of segmentation faults or unexpected
2354 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2359 Many settings can't be controlled through command line options yet.
2367 Fabian Keil <fk@fabiankeil.de>