X-Git-Url: http://git.hungrycats.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=hacks%2Fwebcollage;h=63353a9aef645fdd2fe2066015f01621e358ca65;hb=4361b69d3178d7fc98d0388f9a223af6c2651aba;hp=9d76d1ab64d41c7bb37e63054c6e6c43edf81a08;hpb=c8c6deae79b408cffbc88043c766b3bc12cf0f13;p=xscreensaver diff --git a/hacks/webcollage b/hacks/webcollage index 9d76d1ab..63353a9a 100755 --- a/hacks/webcollage +++ b/hacks/webcollage @@ -1,6 +1,6 @@ #!/usr/bin/perl -w # -# webcollage, Copyright (c) 1999-2005 by Jamie Zawinski +# webcollage, Copyright © 1999-2016 by Jamie Zawinski # This program decorates the screen with random images from the web. # One satisfied customer described it as "a nonstop pop culture brainbath." # @@ -15,15 +15,15 @@ # To run this as a display mode with xscreensaver, add this to `programs': # -# webcollage -root -# webcollage -root -filter 'vidwhacker -stdin -stdout' +# webcollage --root +# webcollage --root --filter 'vidwhacker --stdin --stdout' # # -# You can see this in action at http://www.jwz.org/webcollage/ -- +# You can see this in action at https://www.jwz.org/webcollage/ -- # it auto-reloads about once a minute. To make a page similar to # that on your own system, do this: # -# webcollage -size '800x600' -imagemap $HOME/www/webcollage/index +# webcollage --size '800x600' --imagemap $HOME/www/webcollage/index # # # If you have the "driftnet" program installed, webcollage can display a @@ -34,7 +34,7 @@ # Driftnet is available here: http://www.ex-parrot.com/~chris/driftnet/ # Use it like so: # -# webcollage -root -driftnet +# webcollage --root --driftnet # # Driftnet is the Unix implementation of the MacOS "EtherPEG" program. @@ -48,33 +48,59 @@ use strict; #use diagnostics; -use Socket; require Time::Local; require POSIX; use Fcntl ':flock'; # import LOCK_* constants use POSIX qw(strftime); - -use bytes; # Larry can take Unicode and shove it up his ass sideways. - # Perl 5.8.0 causes us to start getting incomprehensible - # errors about UTF-8 all over the place without this. +use LWP::UserAgent; my $progname = $0; $progname =~ s@.*/@@g; -my $version = q{ $Revision: 1.135 $ }; $version =~ s/^[^0-9]+([0-9.]+).*$/$1/; -my $copyright = "WebCollage $version, Copyright (c) 1999-2005" . +my ($version) = ('$Revision: 1.176 $' =~ m/\s(\d[.\d]+)\s/s); +my $copyright = "WebCollage $version, Copyright (c) 1999-2015" . " Jamie Zawinski \n" . - " http://www.jwz.org/webcollage/\n"; + " https://www.jwz.org/webcollage/\n"; + + + +my @search_methods = ( + # Google is rate-limiting us now, so this works ok from + # a short-running screen saver, but not as a batch job. + # I haven't found a workaround. + # + 7, "googlephotos", \&pick_from_google_image_photos, + 5, "googleimgs", \&pick_from_google_images, + 5, "googlenums", \&pick_from_google_image_numbers, + + # So let's try Bing instead. No rate limiting yet! + # + 14, "bingphotos", \&pick_from_bing_image_photos, + 12, "bingimgs", \&pick_from_bing_images, + 11, "bingnums", \&pick_from_bing_image_numbers, + + 21, "flickr_recent", \&pick_from_flickr_recent, + 16, "flickr_random", \&pick_from_flickr_random, + 9, "livejournal", \&pick_from_livejournal_images, + + # I ran out of usable access tokens, May 2017 + # 23, "instagram", \&pick_from_instagram, + # No longer exists, as of Apr 2014 + # 4, "yahoorand", \&pick_from_yahoo_random_link, + # Twitter destroyed their whole API in 2013. + # 0, "twitpic", \&pick_from_twitpic_images, + # 0, "twitter", \&pick_from_twitter_images, -my @search_methods = ( 56, "altavista", \&pick_from_alta_vista_random_link, - 11, "livejournal", \&pick_from_livejournal_images, - 5, "yahoorand", \&pick_from_yahoo_random_link, - 10, "googlephotos", \&pick_from_google_image_photos, - 5, "googleimgs", \&pick_from_google_images, - 3, "googlenums", \&pick_from_google_image_numbers, - 2, "flickr_recent", \&pick_from_flickr_recent, - 8, "flickr_random", \&pick_from_flickr_random, + # This is a cute way to search for a certain webcams. + # Not included in default methods, since these images + # aren't terribly interesting by themselves. + # See also "SurveillanceSaver". + # + 0, "securitycam", \&pick_from_security_camera, + + # Nonfunctional as of June 2011. + # 0, "altavista", \&pick_from_alta_vista_random_link, # In Apr 2002, Google asked me to stop searching them. # I asked them to add a "random link" url. They said @@ -87,31 +113,31 @@ my @search_methods = ( 56, "altavista", \&pick_from_alta_vista_random_link, # it's no longer possible to do "or" searches on news # images, so we rarely get any hits there any more. # - # 0, "yahoonews", \&pick_from_yahoo_news_text, + # 0, "yahoonews", \&pick_from_yahoo_news_text, # Dec 2004: the ircimages guy's server can't take the # heat, so he started banning the webcollage user agent. # I tried to convince him to add a lighter-weight page to # support webcollage better, but he doesn't care. # - # 0, "ircimages", \&pick_from_ircimages, + # 0, "ircimages", \&pick_from_ircimages, # Dec 2002: Alta Vista has a new "random link" URL now. # They added it specifically to better support webcollage! # That was super cool of them. This is how we used to do # it, before: # - # 0, "avimages", \&pick_from_alta_vista_images, - # 0, "avtext", \&pick_from_alta_vista_text, + # 0, "avimages", \&pick_from_alta_vista_images, + # 0, "avtext", \&pick_from_alta_vista_text, # This broke in 2004. Eh, Lycos sucks anyway. # - # 0, "lycos", \&pick_from_lycos_text, + # 0, "lycos", \&pick_from_lycos_text, # This broke in 2003, I think. I suspect Hotbot is # actually the same search engine data as Lycos. # - # 0, "hotbot", \&pick_from_hotbot_text, + # 0, "hotbot", \&pick_from_hotbot_text, ); # programs we can use to write to the root window (tried in ascending order.) @@ -181,6 +207,11 @@ my %poisoners = ( # (I don't see how they did it, though!) "alwayshotels.com" => 1, # Poisoned Lycos pretty heavily. "nextag.com" => 1, # Poisoned Alta Vista real good. + "ghettodriveby.com" => 1, # Poisoned Google Images. + "crosswordsolver.org" => 1, # Poisoned Google Images. + "xona.com" => 1, # Poisoned Google Images. + "freepatentsonline.com" => 1, # Poisoned Google Images. + "herbdatanz.com" => 1, # Poisoned Google Images. ); @@ -191,14 +222,31 @@ my %poisoners = ( # site" diagnostic message. # my %warningless_sites = ( - "home.earthlink.net" => 1, # Lots of home pages here. - "www.geocities.com" => 1, + "home.earthlink.net" => 1, "www.angelfire.com" => 1, "members.aol.com" => 1, "img.photobucket.com" => 1, "pics.livejournal.com" => 1, "tinypic.com" => 1, "flickr.com" => 1, + "staticflickr.com" => 1, + "pbase.com" => 1, + "blogger.com" => 1, + "multiply.com" => 1, + "wikimedia.org" => 1, + "twitpic.com" => 1, + "amazonaws.com" => 1, + "blogspot.com" => 1, + "photoshelter.com" => 1, + "myspacecdn.com" => 1, + "feedburner.com" => 1, + "wikia.com" => 1, + "ljplus.ru" => 1, + "yandex.ru" => 1, + "imgur.com" => 1, + "yfrog.com" => 1, + "cdninstagram.com" => 1, + "encrypted-tbn3.gstatic.com" => 1, "yimg.com" => 1, # This is where dailynews.yahoo.com stores "eimg.com" => 1, # its images, so pick_from_yahoo_news_text() @@ -217,32 +265,53 @@ my %warningless_sites = ( # my %entity_table = ( "apos" => '\'', - "quot" => '"', "amp" => '&', "lt" => '<', "gt" => '>', - "nbsp" => ' ', "iexcl" => '¡', "cent" => '¢', "pound" => '£', - "curren" => '¤', "yen" => '¥', "brvbar" => '¦', "sect" => '§', - "uml" => '¨', "copy" => '©', "ordf" => 'ª', "laquo" => '«', - "not" => '¬', "shy" => '­', "reg" => '®', "macr" => '¯', - "deg" => '°', "plusmn" => '±', "sup2" => '²', "sup3" => '³', - "acute" => '´', "micro" => 'µ', "para" => '¶', "middot" => '·', - "cedil" => '¸', "sup1" => '¹', "ordm" => 'º', "raquo" => '»', - "frac14" => '¼', "frac12" => '½', "frac34" => '¾', "iquest" => '¿', - "Agrave" => 'À', "Aacute" => 'Á', "Acirc" => 'Â', "Atilde" => 'Ã', - "Auml" => 'Ä', "Aring" => 'Å', "AElig" => 'Æ', "Ccedil" => 'Ç', - "Egrave" => 'È', "Eacute" => 'É', "Ecirc" => 'Ê', "Euml" => 'Ë', - "Igrave" => 'Ì', "Iacute" => 'Í', "Icirc" => 'Î', "Iuml" => 'Ï', - "ETH" => 'Ð', "Ntilde" => 'Ñ', "Ograve" => 'Ò', "Oacute" => 'Ó', - "Ocirc" => 'Ô', "Otilde" => 'Õ', "Ouml" => 'Ö', "times" => '×', - "Oslash" => 'Ø', "Ugrave" => 'Ù', "Uacute" => 'Ú', "Ucirc" => 'Û', - "Uuml" => 'Ü', "Yacute" => 'Ý', "THORN" => 'Þ', "szlig" => 'ß', - "agrave" => 'à', "aacute" => 'á', "acirc" => 'â', "atilde" => 'ã', - "auml" => 'ä', "aring" => 'å', "aelig" => 'æ', "ccedil" => 'ç', - "egrave" => 'è', "eacute" => 'é', "ecirc" => 'ê', "euml" => 'ë', - "igrave" => 'ì', "iacute" => 'í', "icirc" => 'î', "iuml" => 'ï', - "eth" => 'ð', "ntilde" => 'ñ', "ograve" => 'ò', "oacute" => 'ó', - "ocirc" => 'ô', "otilde" => 'õ', "ouml" => 'ö', "divide" => '÷', - "oslash" => 'ø', "ugrave" => 'ù', "uacute" => 'ú', "ucirc" => 'û', - "uuml" => 'ü', "yacute" => 'ý', "thorn" => 'þ', "yuml" => 'ÿ', - "ndash" => '-', "mdash" => "--" + "quot" => '"', "amp" => '&', "lt" => '<', + "gt" => '>', "nbsp" => ' ', "iexcl" => '', + "cent" => "\xA2", "pound" => "\xA3", "curren" => "\xA4", + "yen" => "\xA5", "brvbar" => "\xA6", "sect" => "\xA7", + "uml" => "\xA8", "copy" => "\xA9", "ordf" => "\xAA", + "laquo" => "\xAB", "not" => "\xAC", "shy" => "\xAD", + "reg" => "\xAE", "macr" => "\xAF", "deg" => "\xB0", + "plusmn" => "\xB1", "sup2" => "\xB2", "sup3" => "\xB3", + "acute" => "\xB4", "micro" => "\xB5", "para" => "\xB6", + "middot" => "\xB7", "cedil" => "\xB8", "sup1" => "\xB9", + "ordm" => "\xBA", "raquo" => "\xBB", "frac14" => "\xBC", + "frac12" => "\xBD", "frac34" => "\xBE", "iquest" => "\xBF", + "Agrave" => "\xC0", "Aacute" => "\xC1", "Acirc" => "\xC2", + "Atilde" => "\xC3", "Auml" => "\xC4", "Aring" => "\xC5", + "AElig" => "\xC6", "Ccedil" => "\xC7", "Egrave" => "\xC8", + "Eacute" => "\xC9", "Ecirc" => "\xCA", "Euml" => "\xCB", + "Igrave" => "\xCC", "Iacute" => "\xCD", "Icirc" => "\xCE", + "Iuml" => "\xCF", "ETH" => "\xD0", "Ntilde" => "\xD1", + "Ograve" => "\xD2", "Oacute" => "\xD3", "Ocirc" => "\xD4", + "Otilde" => "\xD5", "Ouml" => "\xD6", "times" => "\xD7", + "Oslash" => "\xD8", "Ugrave" => "\xD9", "Uacute" => "\xDA", + "Ucirc" => "\xDB", "Uuml" => "\xDC", "Yacute" => "\xDD", + "THORN" => "\xDE", "szlig" => "\xDF", "agrave" => "\xE0", + "aacute" => "\xE1", "acirc" => "\xE2", "atilde" => "\xE3", + "auml" => "\xE4", "aring" => "\xE5", "aelig" => "\xE6", + "ccedil" => "\xE7", "egrave" => "\xE8", "eacute" => "\xE9", + "ecirc" => "\xEA", "euml" => "\xEB", "igrave" => "\xEC", + "iacute" => "\xED", "icirc" => "\xEE", "iuml" => "\xEF", + "eth" => "\xF0", "ntilde" => "\xF1", "ograve" => "\xF2", + "oacute" => "\xF3", "ocirc" => "\xF4", "otilde" => "\xF5", + "ouml" => "\xF6", "divide" => "\xF7", "oslash" => "\xF8", + "ugrave" => "\xF9", "uacute" => "\xFA", "ucirc" => "\xFB", + "uuml" => "\xFC", "yacute" => "\xFD", "thorn" => "\xFE", + "yuml" => "\xFF", + + # HTML 4 entities that do not have 1:1 Latin1 mappings. + "bull" => "*", "hellip"=> "...", "prime" => "'", "Prime" => "\"", + "frasl" => "/", "trade" => "[tm]", "larr" => "<-", "rarr" => "->", + "harr" => "<->", "lArr" => "<=", "rArr" => "=>", "hArr" => "<=>", + "empty" => "\xD8", "minus" => "-", "lowast"=> "*", "sim" => "~", + "cong" => "=~", "asymp" => "~", "ne" => "!=", "equiv" => "==", + "le" => "<=", "ge" => ">=", "lang" => "<", "rang" => ">", + "loz" => "<>", "OElig" => "OE", "oelig" => "oe", "Yuml" => "Y", + "circ" => "^", "tilde" => "~", "ensp" => " ", "emsp" => " ", + "thinsp"=> " ", "ndash" => "-", "mdash" => "--", "lsquo" => "`", + "rsquo" => "'", "sbquo" => "'", "ldquo" => "\"", "rdquo" => "\"", + "bdquo" => "\"", "lsaquo"=> "<", "rsaquo"=> ">", ); @@ -320,169 +389,81 @@ sub get_document_1($$$) { if (!defined($timeout)) { $timeout = $http_timeout; } if ($timeout > $http_timeout) { $timeout = $http_timeout; } - if ($timeout <= 0) { - LOG (($verbose_net || $verbose_load), "timed out for $url"); - return (); - } + my $user_agent = "$progname/$version"; + + if ($url =~ m@^https?://www\.altavista\.com/@s || + $url =~ m@^https?://random\.yahoo\.com/@s || + $url =~ m@^https?://[^./]+\.google\.com/@s || + $url =~ m@^https?://www\.livejournal\.com/@s) { + # block this, you turkeys. + $user_agent = 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.7)' . + ' Gecko/20070914 Firefox/2.0.0.7'; + } + + my $ua = LWP::UserAgent->new ( agent => $user_agent, + keep_alive => 0, + env_proxy => 0, + ); + $ua->proxy ('http', $http_proxy) if $http_proxy; + $ua->default_header ('Referer' => $referer) if $referer; + $ua->default_header ('Accept' => '*/*'); + $ua->timeout($timeout) if $timeout; + + if (0) { + $ua->add_handler ("request_send", + sub($$$) { + my ($req, $ua, $h) = @_; + print "\n>>[[\n"; $req->dump; print "\n]]\n"; + return; + }); + $ua->add_handler ("response_data", + sub($$$$) { + my ($req, $ua, $h, $data) = @_; + #print "\n<<[[\n"; print $data; print "\n]]\n"; + return 1; + }); + $ua->add_handler ("request_done", + sub($$$) { + my ($req, $ua, $h) = @_; + print "\n<<[[\n"; $req->dump; print "\n]]\n"; + return; + }); + } + + if ($verbose_http) { + LOG (1, " ==> GET $url"); + LOG (1, " ==> User-Agent: $user_agent"); + LOG (1, " ==> Referer: $referer") if $referer; + } + + my $res = $ua->get ($url); + + my $http = ($res ? $res->status_line : '') || ''; + my $head = ($res ? $res->headers() : '') || ''; + $head = $head->as_string() if $head; + my $body = ($res && $res->is_success ? $res->decoded_content : '') || ''; LOG ($verbose_net, "get_document_1 $url " . ($referer ? $referer : "")); - if (! ($url =~ m@^http://@i)) { - LOG ($verbose_net, "not an HTTP URL: $url"); - return (); - } - - my ($url_proto, $dummy, $serverstring, $path) = split(/\//, $url, 4); - $path = "" unless $path; - - if (!$url_proto || !$serverstring) { - LOG (($verbose_net || $verbose_load), "unparsable URL: $url"); - return (); + $head =~ s/\r\n/\n/gs; + $head =~ s/\r/\n/gs; + if ($verbose_http) { + foreach (split (/\n/, $head)) { + LOG ($verbose_http, " <== $_"); + } } - my ($them,$port) = split(/:/, $serverstring); - $port = 80 unless $port; - - my $them2 = $them; - my $port2 = $port; - if ($http_proxy) { - $serverstring = $http_proxy if $http_proxy; - $serverstring =~ s@^[a-z]+://@@; - ($them2,$port2) = split(/:/, $serverstring); - $port2 = 80 unless $port2; - } + my @L = split(/\r\n|\r|\n/, $body); + my $lines = @L; + LOG ($verbose_http, + " <== [ body ]: $lines lines, " . length($body) . " bytes"); - my ($remote, $iaddr, $paddr, $proto, $line); - $remote = $them2; - if ($port2 =~ /\D/) { $port2 = getservbyname($port2, 'tcp') } - if (!$port2) { - LOG (($verbose_net || $verbose_load), "unrecognised port in $url"); + if (!$http) { + LOG (($verbose_net || $verbose_load), "null response: $url"); return (); } - $iaddr = inet_aton($remote); - if (!$iaddr) { - LOG (($verbose_net || $verbose_load), "host not found: $remote"); - return (); - } - $paddr = sockaddr_in($port2, $iaddr); - - - my $head = ""; - my $body = ""; - - @_ = - eval { - local $SIG{ALRM} = sub { - LOG (($verbose_net || $verbose_load), "timed out ($timeout) for $url"); - die "alarm\n"; - }; - alarm $timeout; - - $proto = getprotobyname('tcp'); - if (!socket(S, PF_INET, SOCK_STREAM, $proto)) { - LOG (($verbose_net || $verbose_load), "socket: $!"); - return (); - } - if (!connect(S, $paddr)) { - LOG (($verbose_net || $verbose_load), "connect($serverstring): $!"); - return (); - } - - select(S); $| = 1; select(STDOUT); - - my $cookie = $cookies{$them}; - - my $user_agent = "$progname/$version"; - - if ($url =~ m@^http://www\.altavista\.com/@ || - $url =~ m@^http://random\.yahoo\.com/@ || - $url =~ m@^http://images\.google\.com/@) { - # block this, you turkeys. - $user_agent = "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.5)" . - " Gecko/20041111 Firefox/1.0"; - - # 28-Jun-2007: Google Images now emits the entire page in JS if - # you claim to be Gecko. They also still block "webcollage". - # They serve non-JS for unrecognised agents, so let's try this... - $user_agent = "NoJavascriptPlease/1.0" - if ($url =~ m@^http://images\.google\.com/@); - } - - my $hdrs = "GET " . ($http_proxy ? $url : "/$path") . " HTTP/1.0\r\n" . - "Host: $them\r\n" . - "User-Agent: $user_agent\r\n"; - if ($referer) { - $hdrs .= "Referer: $referer\r\n"; - } - if ($cookie) { - my @cc = split(/\r?\n/, $cookie); - $hdrs .= "Cookie: " . join('; ', @cc) . "\r\n"; - } - $hdrs .= "\r\n"; - - foreach (split('\r?\n', $hdrs)) { - LOG ($verbose_http, " ==> $_"); - } - print S $hdrs; - my $http = || ""; - - # Kludge: the Yahoo Random Link is now returning as its first - # line "Status: 301" instead of "HTTP/1.0 301 Found". Fix it... - # - $http =~ s@^Status:\s+(\d+)\b@HTTP/1.0 $1@i; - - $_ = $http; - s/[\r\n]+$//s; - LOG ($verbose_http, " <== $_"); - - while () { - $head .= $_; - s/[\r\n]+$//s; - last if m@^$@; - LOG ($verbose_http, " <== $_"); - - if (m@^Set-cookie:\s*([^;\r\n]+)@i) { - set_cookie($them, $1) - } - } - - my $lines = 0; - while () { - $body .= $_; - $lines++; - } - - LOG ($verbose_http, - " <== [ body ]: $lines lines, " . length($body) . " bytes"); - close S; - - if (!$http) { - LOG (($verbose_net || $verbose_load), "null response: $url"); - return (); - } - - $SIG{ALRM} = 'DEFAULT'; # seem to be suffering a race? - return ( $http, $head, $body ); - }; - die if ($@ && $@ ne "alarm\n"); # propagate errors - - if ($@ && $@ ne "alarm\n") { - print STDERR blurb() . "DIE " . join(" ", $@) . "\n"; - die; - } - - if ($@) { - # timed out - $head = undef; - $body = undef; - $suppress_audit = 1; - return (); - } else { - # didn't - alarm 0; - return @_; - } + return ( $http, $head, $body ); } @@ -537,12 +518,12 @@ sub get_document($$;$) { $url = $location; if ($url =~ m@^/@) { - $referer =~ m@^(http://[^/]+)@i; + $referer =~ m@^(https?://[^/]+)@i; $url = $1 . $url; } elsif (! ($url =~ m@^[a-z]+:@i)) { $_ = $referer; - s@[^/]+$@@g if m@^http://[^/]+/@i; - $_ .= "/" if m@^http://[^/]+$@i; + s@[^/]+$@@g if m@^https?://[^/]+/@i; + $_ .= "/" if m@^https?://[^/]+$@i; $url = $_ . $url; } @@ -620,12 +601,12 @@ sub pick_image_from_body($$) { # if there's at least one slash after the host, take off the last # pathname component - if ( m@^http://[^/]+/@io ) { + if ( m@^https?://[^/]+/@io ) { $base =~ s@[^/]+$@@go; } # if there are no slashes after the host at all, put one on the end. - if ( m@^http://[^/]+$@io ) { + if ( m@^https?://[^/]+$@io ) { $base .= "/"; } @@ -684,37 +665,46 @@ sub pick_image_from_body($$) { my %unique_urls; foreach (split(/ * 1000) { - LOG (($verbose_filter || $verbose_load), - "excessive keywords ($L bytes) in $url: rejecting."); - $rejected_urls{$url} = $L; - $body = undef; - $_ = undef; - return (); - } else { - LOG ($verbose_filter, " keywords ($L bytes) in $url (ok)"); - } + my $L = length($_); + if ($L > 1000) { + LOG (($verbose_filter || $verbose_load), + "excessive keywords ($L bytes) in $url: rejecting."); + $rejected_urls{$url} = $L; + $body = undef; + $_ = undef; + return (); + } else { + LOG ($verbose_filter, " keywords ($L bytes) in $url (ok)"); } - } elsif ( m/^(img|a) .*(src|href) ?= ?\"? ?(.*?)[ >\"]/io ) { + } elsif (m/^ (IMG|A) \b .* (SRC|HREF) \s* = \s* ["']? (.*?) [ "'<>] /six || + m/^ (LINK|META) \b .* (REL|PROPERTY) \s* = \s* + ["']? (image_src|og:image) ["']? /six) { - my $was_inline = (! ( "$1" eq "a" || "$1" eq "A" )); + my $was_inline = (lc($1) eq 'img'); + my $was_meta = (lc($1) eq 'link' || lc($1) eq 'meta'); my $link = $3; + + # For + # and + # + if ($was_meta) { + next unless (m/ (HREF|CONTENT) \s* = \s* ["']? (.*?) [ "'<>] /six); + $link = $2; + } + my ( $width ) = m/width ?=[ \"]*(\d+)/oi; my ( $height ) = m/height ?=[ \"]*(\d+)/oi; $_ = $link; if ( m@^/@o ) { my $site; - ( $site = $base ) =~ s@^(http://[^/]*).*@$1@gio; + ( $site = $base ) =~ s@^(https?://[^/]*).*@$1@gio; $_ = "$site$link"; } elsif ( ! m@^[^/:?]+:@ ) { $_ = "$base$link"; @@ -723,7 +713,7 @@ sub pick_image_from_body($$) { } # skip non-http - if ( ! m@^http://@io ) { + if ( ! m@^https?://@io ) { next; } @@ -779,20 +769,24 @@ sub pick_image_from_body($$) { LOG ($verbose_filter, " image $url" . ($width && $height ? " (${width}x${height})" : "") . - ($was_inline ? " (inline)" : "")); + ($was_meta ? " (meta)" : $was_inline ? " (inline)" : "")); - $urls[++$#urls] = $url; - $unique_urls{$url}++; - # JPEGs are preferable to GIFs and PNGs. - $_ = $url; - if ( ! m@[.](gif|png)$@io ) { - $urls[++$#urls] = $url; + my $weight = 1; + + if ($was_meta) { + $weight = 20; # meta tag images are far preferable to inline images. + } else { + if ($url !~ m@[.](gif|png)$@io ) { + $weight += 2; # JPEGs are preferable to GIFs and PNGs. + } + if (! $was_inline) { + $weight += 4; # pointers to images are preferable to inlined images. + } } - # pointers to images are preferable to inlined images. - if ( ! $was_inline ) { - $urls[++$#urls] = $url; + $unique_urls{$url}++; + for (my $i = 0; $i < $weight; $i++) { $urls[++$#urls] = $url; } } @@ -819,6 +813,7 @@ sub pick_image_from_body($$) { return $url; } + # Given a URL and the RSS feed from that URL, pick a random image from # the feed. This is a lot simpler than extracting images out of a page: # we already know we have reasonable images, so we just pick one. @@ -854,7 +849,10 @@ sub pick_image_from_rss($$) { sub pick_dictionary() { my @dicts = ("/usr/dict/words", "/usr/share/dict/words", - "/usr/share/lib/dict/words"); + "/usr/share/lib/dict/words", + "/usr/share/dict/cracklib-small", + "/usr/share/dict/cracklib-words" + ); foreach my $f (@dicts) { if (-f $f) { $wordlist = $f; @@ -868,12 +866,9 @@ sub pick_dictionary() { # sub random_word() { - local *IN; - if (! open (IN, "<$wordlist")) { - return undef; - } + return undef unless open (my $in, '<', $wordlist); - my $size = (stat(IN))[7]; + my $size = (stat($in))[7]; my $word = undef; my $count = 0; @@ -882,9 +877,9 @@ sub random_word() { if (++$count > 100); my $pos = int (rand ($size)); - if (seek (IN, $pos, 0)) { - $word = ; # toss partial line - $word = ; # keep next line + if (seek ($in, $pos, 0)) { + $word = <$in>; # toss partial line + $word = <$in>; # keep next line } next unless ($word); @@ -908,7 +903,7 @@ sub random_word() { last if ($word); } - close (IN); + close ($in); if ( $word =~ s/\s/\+/gs ) { # convert intra-word spaces to "+". $word = "\%22$word\%22"; # And put quotes (%22) around it. @@ -919,13 +914,12 @@ sub random_word() { sub random_words($) { - my ($or_p) = @_; - my $sep = ($or_p ? "%20OR%20" : "%20"); - return (random_word . $sep . - random_word . $sep . - random_word . $sep . - random_word . $sep . - random_word); + my ($sep) = @_; + return (random_word() . $sep . + random_word() . $sep . + random_word() . $sep . + random_word() . $sep . + random_word()); } @@ -1025,11 +1019,10 @@ sub pick_from_search_engine($$$) { 1 while ($search_count =~ s/^(\d+)(\d{3})/$1,$2/); # if ($search_count eq "?" || $search_count eq "0") { -# local *OUT; # my $file = "/tmp/wc.html"; -# open(OUT, ">$file") || error ("writing $file: $!"); -# print OUT $body; -# close OUT; +# open (my $out, '>', $file) || error ("writing $file: $!"); +# print $out $body; +# close $out; # print STDERR blurb() . "###### wrote $file\n"; # } @@ -1045,11 +1038,22 @@ sub pick_from_search_engine($$$) { s/(]+)>@i; + my ($u) = m@]+)@i; next unless $u; + my ($u2) = m@]+)@i; - if ($u =~ m/^\"([^\"]*)\"/) { $u = $1; } # quoted string - elsif ($u =~ m/^([^\s]*)\s/) { $u = $1; } # or token + if (m/\bm="\{(.*?)\}"/s) { # Bing info is inside JSON crud + my $json = html_unquote($1); + my ($href) = ($json =~ m/\b(?:surl|purl)\"?:\s*"(.*?)"/s); + my ($img) = ($json =~ m/\b(?:imgurl|murl)\"?:\s*"(.*?)"/s); + $u = "$img\t$href" if ($img && $href); + + } elsif ($u2 && $u2 =~ m@://[^/]*\.gstatic\.com/@s) { $u = $u2; + $u =~ s/^\"|\"$//s; + + } elsif ($u =~ m/^\"([^\"]*)\"/) { $u = $1 # quoted string + } elsif ($u =~ m/^([^\s]*)\s/) { $u = $1; # or token + } if ( $rejected_urls{$u} ) { LOG ($verbose_filter, " pre-rejecting candidate: $u"); @@ -1077,7 +1081,7 @@ sub depoison(@) { my (@urls) = @_; my @urls2 = (); foreach (@urls) { - my ($h) = m@^http://([^/: \t\r\n]+)@i; + my ($h) = m@^https?://([^/: \t\r\n]+)@i; next unless defined($h); @@ -1143,52 +1147,53 @@ sub pick_image_from_pages($$$$@) { } -############################################################################ +############################################################################# +## +## Pick images from random pages returned by the Yahoo Random Link +## +############################################################################# +# +## yahoorand +#my $yahoo_random_link = "http://random.yahoo.com/fast/ryl"; # -# Pick images from random pages returned by the Yahoo Random Link # -############################################################################ - -# yahoorand -my $yahoo_random_link = "http://random.yahoo.com/fast/ryl"; - - # Picks a random page; picks a random image on that page; # returns two URLs: the page containing the image, and the image. # Returns () if nothing found this time. # -sub pick_from_yahoo_random_link($) { - my ($timeout) = @_; - - print STDERR "\n\n" if ($verbose_load); - LOG ($verbose_load, "URL: $yahoo_random_link"); - - $last_search = $yahoo_random_link; # for warnings - - $suppress_audit = 1; - - my ( $base, $body ) = get_document ($yahoo_random_link, undef, $timeout); - if (!$base || !$body) { - $body = undef; - return; - } - - LOG ($verbose_load, "redirected to: $base"); - - my $img = pick_image_from_body ($base, $body); - $body = undef; - - if ($img) { - return ($base, $img); - } else { - return (); - } -} +#sub pick_from_yahoo_random_link($) { +# my ($timeout) = @_; +# +# print STDERR "\n\n" if ($verbose_load); +# LOG ($verbose_load, "URL: $yahoo_random_link"); +# +# $last_search = $yahoo_random_link; # for warnings +# +# $suppress_audit = 1; +# +# my ( $base, $body ) = get_document ($yahoo_random_link, undef, $timeout); +# if (!$base || !$body) { +# $body = undef; +# return; +# } +# +# LOG ($verbose_load, "redirected to: $base"); +# +# my $img = pick_image_from_body ($base, $body); +# $body = undef; +# +# if ($img) { +# return ($base, $img); +# } else { +# return (); +# } +#} ############################################################################ # # Pick images from random pages returned by the Alta Vista Random Link +# Note: this seems to have gotten a *lot* less random lately (2007). # ############################################################################ @@ -1269,7 +1274,7 @@ sub pick_from_alta_vista_images($) { next unless ($u =~ s/^.*\*\*(http%3a.*$)/$1/gsi); $u = url_unquote($u); - next unless ($u =~ m@^http://@i); # skip non-HTTP or relative URLs + next unless ($u =~ m@^https?://@i); # skip non-HTTP or relative URLs next if ($u =~ m@[/.]altavista\.com\b@i); # skip altavista builtins next if ($u =~ m@[/.]yahoo\.com\b@i); # yahoo and av in cahoots? next if ($u =~ m@[/.]doubleclick\.net\b@i); # you cretins @@ -1290,55 +1295,41 @@ sub pick_from_alta_vista_images($) { ############################################################################ # -# Pick images by feeding random words into Google Image Search. -# By Charles Gales +# Pick images from Aptix security cameras +# Cribbed liberally from google image search code. +# By Jason Sullivan # ############################################################################ +my $aptix_images_url = ("http://www.google.com/search" . + "?q=inurl:%22jpg/image.jpg%3Fr%3D%22"); -my $google_images_url = "http://images.google.com/images" . - "?site=images" . # photos - "&btnG=Search" . # graphics - "&safe=off" . # no screening - "&imgsafe=off" . - "&q="; - -# googleimgs -sub pick_from_google_images($;$$) { - my ($timeout, $words, $max_page) = @_; - - if (!defined($words)) { - $words = random_word; # only one word for Google - } +# securitycam +sub pick_from_security_camera($) { + my ($timeout) = @_; my $page = (int(rand(9)) + 1); - my $num = 20; # 20 images per page - my $search_url = $google_images_url . $words; + my $num = 20; # 20 images per page + my $search_url = $aptix_images_url; if ($page > 1) { $search_url .= "&start=" . $page*$num; # page number - $search_url .= "&num=" . $num; #images per page + $search_url .= "&num=" . $num; #images per page } my ($search_hit_count, @subpages) = - pick_from_search_engine ($timeout, $search_url, $words); + pick_from_search_engine ($timeout, $search_url, ''); my @candidates = (); my %referers; foreach my $u (@subpages) { - next unless ($u =~ m@imgres\?imgurl@i); # All pics start with this - next if ($u =~ m@[/.]google\.com\b@i); # skip google builtins + next if ($u =~ m@[/.]google\.com\b@i); # skip google builtins (most links) + next unless ($u =~ m@jpg/image.jpg\?r=@i); # All pics contain this - if ($u =~ m@^/imgres\?imgurl=(.*?)\&imgrefurl=(.*?)\&@) { - my $ref = $2; - my $img = $1; - $img = "http://$img" unless ($img =~ m/^http:/i); - - LOG ($verbose_filter, " candidate: $ref"); - push @candidates, $img; - $referers{$img} = $ref; + LOG ($verbose_filter, " candidate: $u"); + push @candidates, $u; + $referers{$u} = $u; } - } @candidates = depoison (@candidates); return () if ($#candidates < 0); @@ -1350,6 +1341,49 @@ sub pick_from_google_images($;$$) { return ($ref, $img); } + +############################################################################ +# +# Pick images by feeding random words into Google Image Search. +# By Charles Gales +# +############################################################################ + +my $google_images_url = 'https://www.google.com/search' . + '?source=lnms&tbm=isch&tbs=isz:l&q='; + +# googleimgs +sub pick_from_google_images($;$$) { + my ($timeout, $words, $max_page) = @_; + + if (!defined($words)) { + $words = random_word(); # only one word for Google + } + + my $off = int(rand(40)); + my $search_url = $google_images_url . $words . "&start=" . $off; + + my ($search_hit_count, @subpages) = + pick_from_search_engine ($timeout, $search_url, $words); + + my @candidates = (); + foreach my $u (@subpages) { + $u = html_unquote($u); + # next if ($u =~ m@^https?://[^.]*\.(google|youtube)\.com/@s); + next unless ($u =~ m@^https?://[^/]*\.gstatic\.com@s); + LOG ($verbose_filter, " candidate: $u"); + push @candidates, $u; + } + + @candidates = depoison (@candidates); + return () if ($#candidates < 0); + my $i = int(rand($#candidates+1)); + my $img = $candidates[$i]; + + LOG ($verbose_load, "picked image " . ($i+1) . ": $img"); + return ($img, $img); +} + ############################################################################ @@ -1381,6 +1415,7 @@ sub pick_from_google_image_numbers($) { # Google Image Search. # By jwz, inspired by the excellent Random Personal Picture Finder # at http://www.diddly.com/random/ +# May 2017: Commented out a bunch of formats that have fallen out of favor. # ############################################################################ @@ -1389,32 +1424,32 @@ my @photomakers = ( # Common digital camera file name formats, as described at # http://www.diddly.com/random/about.html # - sub { sprintf ("dcp%05d.jpg", int(rand(4000))); }, # Kodak +# sub { sprintf ("dcp%05d.jpg", int(rand(4000))); }, # Kodak sub { sprintf ("dsc%05d.jpg", int(rand(4000))); }, # Nikon sub { sprintf ("dscn%04d.jpg", int(rand(4000))); }, # Nikon - sub { sprintf ("mvc-%03d.jpg", int(rand(999))); }, # Sony Mavica - sub { sprintf ("mvc%05d.jpg", int(rand(9999))); }, # Sony Mavica - sub { sprintf ("P101%04d.jpg", int(rand(9999))); }, # Olympus w/ date=101 - sub { sprintf ("P%x%02d%04d.jpg", # Olympus - int(rand(0xC)), int(rand(30))+1, - rand(9999)); }, +# sub { sprintf ("mvc-%03d.jpg", int(rand(999))); }, # Sony Mavica +# sub { sprintf ("mvc%05d.jpg", int(rand(9999))); }, # Sony Mavica +# sub { sprintf ("P101%04d.jpg", int(rand(9999))); }, # Olympus w/ date=101 +# sub { sprintf ("P%x%02d%04d.jpg", # Olympus +# int(rand(0xC)), int(rand(30))+1, +# rand(9999)); }, sub { sprintf ("IMG_%03d.jpg", int(rand(999))); }, # ? - sub { sprintf ("IMAG%04d.jpg", int(rand(9999))); }, # RCA and Samsung - sub { my $n = int(rand(9999)); # Canon - sprintf ("1%02d-%04d.jpg", int($n/100), $n); }, - sub { my $n = int(rand(9999)); # Canon - sprintf ("1%02d-%04d_IMG.jpg", - int($n/100), $n); }, +# sub { sprintf ("IMAG%04d.jpg", int(rand(9999))); }, # RCA and Samsung +# sub { my $n = int(rand(9999)); # Canon +# sprintf ("1%02d-%04d.jpg", int($n/100), $n); }, +# sub { my $n = int(rand(9999)); # Canon +# sprintf ("1%02d-%04d_IMG.jpg", +# int($n/100), $n); }, sub { sprintf ("IMG_%04d.jpg", int(rand(9999))); }, # Canon sub { sprintf ("dscf%04d.jpg", int(rand(9999))); }, # Fuji Finepix - sub { sprintf ("pdrm%04d.jpg", int(rand(9999))); }, # Toshiba PDR - sub { sprintf ("IM%06d.jpg", int(rand(9999))); }, # HP Photosmart - sub { sprintf ("EX%06d.jpg", int(rand(9999))); }, # HP Photosmart +# sub { sprintf ("pdrm%04d.jpg", int(rand(9999))); }, # Toshiba PDR +# sub { sprintf ("IM%06d.jpg", int(rand(9999))); }, # HP Photosmart +# sub { sprintf ("EX%06d.jpg", int(rand(9999))); }, # HP Photosmart # sub { my $n = int(rand(3)); # Kodak DC-40,50,120 # sprintf ("DC%04d%s.jpg", int(rand(9999)), # $n == 0 ? 'S' : $n == 1 ? 'M' : 'L'); }, sub { sprintf ("pict%04d.jpg", int(rand(9999))); }, # Minolta Dimage - sub { sprintf ("P%07d.jpg", int(rand(9999))); }, # Kodak DC290 +# sub { sprintf ("P%07d.jpg", int(rand(9999))); }, # Kodak DC290 # sub { sprintf ("%02d%02d%04d.jpg", # Casio QV3000, QV4000 # int(rand(12))+1, int(rand(31))+1, # int(rand(999))); }, @@ -1423,9 +1458,9 @@ my @photomakers = ( # int(rand(12))+1, int(rand(31))+1, # int(rand(999))); }, sub { sprintf ("IMGP%04d.jpg", int(rand(9999))); }, # Pentax Optio S - sub { sprintf ("PANA%04d.jpg", int(rand(9999))); }, # Panasonic vid still +# sub { sprintf ("PANA%04d.jpg", int(rand(9999))); }, # Panasonic vid still sub { sprintf ("HPIM%04d.jpg", int(rand(9999))); }, # HP Photosmart - sub { sprintf ("PCDV%04d.jpg", int(rand(9999))); }, # ? +# sub { sprintf ("PCDV%04d.jpg", int(rand(9999))); }, # ? ); @@ -1436,12 +1471,96 @@ sub pick_from_google_image_photos($) { my $i = int(rand($#photomakers + 1)); my $fn = $photomakers[$i]; my $file = &$fn; - my $words .= $file . "%20filetype:jpg"; + #$file .= "%20filetype:jpg"; + + pick_from_google_images ($timeout, $file); +} + + +############################################################################ +# +# Pick images by feeding random words into Google Image Search. +# By the way: fuck Microsoft. +# +############################################################################ + +my $bing_images_url = "http://www.bing.com/images/async?q="; + + +# bingimgs +sub pick_from_bing_images($;$$) { + my ($timeout, $words, $max_page) = @_; + + if (!defined($words)) { + $words = random_word(); # only one word for Bing + } + + my $off = int(rand(300)); + my $search_url = $bing_images_url . $words . "&first=" . $off; + + my ($search_hit_count, @subpages) = + pick_from_search_engine ($timeout, $search_url, $words); + + my @candidates = (); + my %referers; + foreach my $u (@subpages) { + my ($img, $ref) = ($u =~ m/^(.*?)\t(.*)$/s); + next unless $img; + LOG ($verbose_filter, " candidate: $ref"); + push @candidates, $img; + $referers{$img} = $ref; + } + + @candidates = depoison (@candidates); + return () if ($#candidates < 0); + my $i = int(rand($#candidates+1)); + my $img = $candidates[$i]; + my $ref = $referers{$img}; - pick_from_google_images ($timeout, $words); + LOG ($verbose_load, "picked image " . ($i+1) . ": $img (on $ref)"); + return ($ref, $img); } + + +############################################################################ +# +# Pick images by feeding random numbers into Bing Image Search. +# +############################################################################ + +# bingnums +sub pick_from_bing_image_numbers($) { + my ($timeout) = @_; + + my $max = 9999; + my $number = int(rand($max)); + + $number = sprintf("%04d", $number) + if (rand() < 0.3); + + pick_from_bing_images ($timeout, "$number"); +} + + +############################################################################ +# +# Pick images by feeding random numbers into Bing Image Search. +# +############################################################################ + +# bingphotos +sub pick_from_bing_image_photos($) { + my ($timeout) = @_; + + my $i = int(rand($#photomakers + 1)); + my $fn = $photomakers[$i]; + my $file = &$fn; + + pick_from_bing_images ($timeout, $file); +} + ############################################################################ # @@ -1462,7 +1581,7 @@ my $alta_vista_url = "http://www.altavista.com/web/results" . sub pick_from_alta_vista_text($) { my ($timeout) = @_; - my $words = random_words(0); + my $words = random_words('%20'); my $page = (int(rand(9)) + 1); my $search_url = $alta_vista_url . $words; @@ -1485,7 +1604,7 @@ sub pick_from_alta_vista_text($) { next unless ($u =~ s/^.*\*\*(http%3a.*$)/$1/gsi); $u = url_unquote($u); - next unless ($u =~ m@^http://@i); # skip non-HTTP or relative URLs + next unless ($u =~ m@^https?://@i); # skip non-HTTP or relative URLs next if ($u =~ m@[/.]altavista\.com\b@i); # skip altavista builtins next if ($u =~ m@[/.]yahoo\.com\b@i); # yahoo and av in cahoots? @@ -1541,7 +1660,7 @@ sub pick_from_hotbot_text($) { # next unless ($u =~ m@/director.asp\?.*\btarget=([^&]+)@); # $u = url_decode($1); - next unless ($u =~ m@^http://@i); # skip non-HTTP or relative URLs + next unless ($u =~ m@^https?://@i); # skip non-HTTP or relative URLs next if ($u =~ m@[/.]hotbot\.com\b@i); # skip hotbot builtins next if ($u =~ m@[/.]lycos\.com\b@i); # skip hotbot builtins next if ($u =~ m@[/.]inktomi\.com\b@i); # skip hotbot builtins @@ -1589,14 +1708,14 @@ sub pick_from_lycos_text($) { # Lycos plays redirection games. # (not any more?) -# next unless ($u =~ m@^http://click.lycos.com/director.asp +# next unless ($u =~ m@^https?://click.lycos.com/director.asp # .* # \btarget=([^&]+) # .* # @x); # $u = url_decode($1); - next unless ($u =~ m@^http://@i); # skip non-HTTP or relative URLs + next unless ($u =~ m@^https?://@i); # skip non-HTTP or relative URLs next if ($u =~ m@[/.]hotbot\.com\b@i); # skip lycos builtins next if ($u =~ m@[/.]lycos\.com\b@i); # skip lycos builtins next if ($u =~ m@[/.]terralycos\.com\b@i); # skip lycos builtins @@ -1639,11 +1758,11 @@ sub pick_from_yahoo_news_text($) { foreach my $u (@subpages) { # de-redirectize the URLs - $u =~ s@^http://rds\.yahoo\.com/.*-http%3A@http:@s; + $u =~ s@^https?://rds\.yahoo\.com/.*-http%3A@http:@s; # only accept URLs on Yahoo's news site - next unless ($u =~ m@^http://dailynews\.yahoo\.com/@i || - $u =~ m@^http://story\.news\.yahoo\.com/@i); + next unless ($u =~ m@^https?://dailynews\.yahoo\.com/@i || + $u =~ m@^https?://story\.news\.yahoo\.com/@i); next unless ($u =~ m@&u=/@); LOG ($verbose_filter, " candidate: $u"); @@ -1680,7 +1799,10 @@ sub pick_from_livejournal_images($) { $last_search = $livejournal_img_url; # for warnings my ( $base, $body ) = get_document ($livejournal_img_url, undef, $timeout); - return () unless $body; + + # Often the document comes back empty. If so, just use the cache. + # return () unless $body; + $body = '' unless defined($body); $body =~ s/\n/ /gs; $body =~ s/(([^<>]*)@si); + my $page = html_unquote ($1); + + $page =~ s@/$@@s; + $page .= '/full'; + + next if ($twitpic_cache{$page}); # already have it + + LOG ($verbose_filter, " candidate: $page"); + push @twitpic_cache, $page; + $twitpic_cache{$page} = $page; + } + } + + # Pull from the cache. + + return () if ($#twitpic_cache == -1); + + my $n = $#twitpic_cache+1; + my $i = int(rand($n)); + my $page = $twitpic_cache[$i]; + + # delete this one from @twitpic_cache and from %twitpic_cache. + # + @twitpic_cache = ( @twitpic_cache[0 .. $i-1], + @twitpic_cache[$i+1 .. $#twitpic_cache] ); + delete $twitpic_cache{$page}; + + # Keep the size of the cache under the limit by nuking older entries + # + while ($#twitpic_cache >= $twitpic_cache_size) { + my $page = shift @twitpic_cache; + delete $twitpic_cache{$page}; + } + + ( $base, $body ) = get_document ($page, undef, $timeout); + my $img = undef; + $body = '' unless defined($body); + + foreach (split (/= $twitter_cache_size) { + my $page = shift @twitter_cache; + delete $twitter_cache{$page}; + } + + LOG ($verbose_load, "picked page $url"); + + $suppress_audit = 1; + + return ($page, $url); +} + ############################################################################ # @@ -1789,7 +2100,7 @@ sub pick_from_ircimages($) { # ############################################################################ -my $flickr_img_url = "http://www.flickr.com/photos/"; +my $flickr_img_url = "http://www.flickr.com/explore/"; # Like LiveJournal, the Flickr page of images tends to update slowly, # so remember the last N entries on it and randomly select from those. @@ -1812,28 +2123,38 @@ sub pick_from_flickr_recent($) { $last_search .= "?start=$start" if ($start > 0); my ( $base, $body ) = get_document ($last_search, undef, $timeout); - return () unless $body; - $body =~ s/[\r\n]/ /gs; - $body =~ s/(]* \b HREF=\"([^<>\"]+)\" [^<>]* > \s* - ]* \b SRC=\"([^<>\"]+)\" @xsi; - next unless defined ($thumb); - $page = html_unquote ($page); - $thumb = html_unquote ($thumb); - next unless ($thumb =~ m@^http://farm\d*\.static\.flickr\.com/@); + if ($body =~ m@{ *"_data": \[ ( .*? \} ) \]@six) { + $body = $1; + } else { + LOG ($verbose_load, "flickr unparsable: $last_search"); + return (); + } + + $body =~ s/[\r\n]/ /gs; + $body =~ s/(\},) *(\{)/$1\n$2/gs; # "_flickrModelRegistry" + + foreach my $chunk (split (/\n/, $body)) { + my ($img) = ($chunk =~ m@"displayUrl": *"(.*?)"@six); + next unless defined ($img); + $img =~ s/\\//gs; + $img = "//" unless ($img =~ m@^/@s); + $img = "http:$img" unless ($img =~ m/^http/s); + + my ($user) = ($chunk =~ m/"pathAlias": *"(.*?)"/si); + next unless defined ($user); - my $base = "http://www.flickr.com/"; - $page =~ s@^/@$base@; - $thumb =~ s@^/@$base@; + my ($id) = ($img =~ m@/\d+/(\d+)_([\da-f]+)_@si); + my ($page) = "https://www.flickr.com/photos/$user/$id/"; - my $img = $thumb; - $img =~ s/_[a-z](\.[a-z\d]+)$/$1/si; # take off "thumb" suffix + # $img =~ s/_[a-z](\.[a-z\d]+)$/$1/si; # take off "thumb" suffix $count++; next if ($flickr_cache{$img}); # already have it @@ -1877,8 +2198,8 @@ sub pick_from_flickr_recent($) { # ############################################################################ -my $flickr_rss_base = ("http://www.flickr.com/services/feeds/photos_public.gne" . - "?format=rss_200_enc&tags="); +my $flickr_rss_base = ("http://www.flickr.com/services/feeds/photos_public.gne". + "?format=rss_200_enc&tagmode=any&tags="); # Picks a random RSS feed; picks a random image from that feed; # returns 2 URLs: the page containing the image, and the image. @@ -1888,10 +2209,15 @@ my $flickr_rss_base = ("http://www.flickr.com/services/feeds/photos_public.gne" sub pick_from_flickr_random($) { my $timeout = shift; - my $rss = $flickr_rss_base . random_word(); + my $words = random_words(','); + my $rss = $flickr_rss_base . $words; $last_search = $rss; + $_ = $words; + s/,/ /g; + print STDERR "\n\n" if ($verbose_load); + LOG ($verbose_load, "words: $_"); LOG ($verbose_load, "URL: $last_search"); $suppress_audit = 1; @@ -1911,6 +2237,77 @@ sub pick_from_flickr_random($) { return ($base, $img); } + +############################################################################ +# +# Pick random images from Instagram. +# +############################################################################ + +my $instagram_url_base = "https://api.instagram.com/v1/media/popular"; + +# instagram_random +sub pick_from_instagram($) { + my $timeout = shift; + + # Liberated access tokens. + # jsdo.it search for: instagram client_id + # Google search for: instagram "&client_id=" site:jsfiddle.net + my @tokens = (#'b59fbe4563944b6c88cced13495c0f49', # gramfeed.com + #'fa26679250df49c48a33fbcf30aae989', # instac.at + #'d9494686198d4dfeb954979a3e270e5e', # iconosquare.com + #'793ef48bb18e4197b61afce2d799b81c', # jsdo.it + #'67b8a3e0073449bba70600d0fc68e6cb', # jsdo.it + #'26a098e0df4d4b9ea8b4ce6c505b7742', # jsdo.it + #'2437cbcd906a4c10940f990d283d3cd5', # jsdo.it + #'191c7d7d5312464cbd92134f36ffdab5', # jsdo.it + #'acfec809437b4340b2c38f66503af774', # jsdo.it + #'e9f77604a3a24beba949c12d18130988', # jsdo.it + #'2cd7bcf68ae346529770073d311575b3', # jsdo.it + #'830c600fe8d742e2ab3f3b94f9bb22b7', # jsdo.it + #'55865a0397ad41e5997dd95ef4df8da1', # jsdo.it + #'192a5742f3644ea8bed1d25e439286a8', # jsdo.it + #'38ed1477e7a44595861b8842cdb8ba23', # jsdo.it + #'e52f79f645f54488ad0cc47f6f55ade6', # jsfiddle.net + ); + + my $tok = $tokens[int(rand($#tokens+1))]; + $last_search = $instagram_url_base . "?client_id=" . $tok; + + print STDERR "\n\n" if ($verbose_load); + LOG ($verbose_load, "URL: $last_search"); + + my ( $base, $body ) = get_document ($last_search, undef, $timeout); + if (!$base || !$body) { + $body = undef; + return; + } + + $body =~ s/("link")/\001$1/gs; + my @chunks = split(/\001/, $body); + shift @chunks; + my @urls = (); + foreach (@chunks) { + s/\\//gs; + my ($url) = m/"link":\s*"(.*?)"/s; + my ($img) = m/"standard_resolution":\{"url":\s*"(.*?)"/s; + ($img) = m/"url":\s*"(.*?)"/s unless $url; + next unless ($url && $img); + push @urls, [ $url, $img ]; + } + + if ($#urls < 0) { + LOG ($verbose_load, "no images on $last_search"); + return (); + } + + my $i = int(rand($#urls+1)); + my ($url, $img) = @{$urls[$i]}; + + LOG ($verbose_load, "picked image " .($i+1) . "/" . ($#urls+1) . ": $url"); + return ($url, $img); +} + ############################################################################ # @@ -1940,16 +2337,15 @@ sub pick_from_driftnet($) { $last_search = $id; while ($now = time, $now < $start + $timeout) { - local *DIR; - opendir (DIR, $dir) || error ("$dir: $!"); - while (my $file = readdir(DIR)) { + opendir (my $dir, $dir) || error ("$dir: $!"); + while (my $file = readdir($dir)) { next if ($file =~ m/^\./); $file = "$dir/$file"; - closedir DIR; + closedir ($dir); LOG ($verbose_load, "picked file $file ($id)"); return ($id, $file); } - closedir DIR; + closedir ($dir); } LOG (($verbose_net || $verbose_load), "timed out for $id"); return (); @@ -1962,15 +2358,14 @@ sub get_driftnet_file($) { error ("\$driftnet_dir unset?") unless ($driftnet_dir); my $id = $driftnet_magic; - my $re = qr/$driftnet_dir/; error ("$id: $file not in $driftnet_dir?") - unless ($file =~ m@^$re@o); + unless ($file =~ m@^\Q$driftnet_dir@o); - local *IN; - open (IN, $file) || error ("$id: $file: $!"); + open (my $in, '<', $file) || error ("$id: $file: $!"); my $body = ''; - while () { $body .= $_; } - close IN || error ("$id: $file: $!"); + local $/ = undef; # read entire file + $body = <$in>; + close ($in) || error ("$id: $file: $!"); unlink ($file) || error ("$id: $file: rm: $!"); return ($id, $body); } @@ -2019,8 +2414,8 @@ sub spawn_driftnet($) { } # local-directory -sub pick_from_local_dir { - my ( $timeout ) = @_; +sub pick_from_local_dir($) { + my ($timeout) = @_; my $id = $local_magic; $last_search = $id; @@ -2033,27 +2428,27 @@ sub pick_from_local_dir { my $v = ($verbose_exec ? "-v" : ""); my $pick = `xscreensaver-getimage-file $v "$dir"`; + $pick =~ s/\s+$//s; + $pick = "$dir/$pick" unless ($pick =~ m@^/@s); # relative path LOG ($verbose_load, "picked file $pick ($id)"); return ($id, $pick); } -sub get_local_file { +sub get_local_file($) { my ($file) = @_; error ("\$local_dir unset?") unless ($local_dir); my $id = $local_magic; - my $re = qr/$local_dir/; error ("$id: $file not in $local_dir?") - unless ($file =~ m@^$re@o); + unless ($file =~ m@^\Q$local_dir@o); - local *IN; - open (IN, $file) || error ("$id: $file: $!"); - my $body = ''; - while () { $body .= $_; } - close IN || error ("$id: $file: $!"); + open (my $in, '<:raw', $file) || error ("$id: $file: $!"); + local $/ = undef; # read entire file + my $body = <$in>; + close ($in) || error ("$id: $file: $!"); return ($id, $body); } @@ -2283,8 +2678,9 @@ sub report_performance() { my $suc = $stats_successes{$name} || 0; my $pct = int($suc * 100 / $try); my $secs = $stats_elapsed{$name}; - my $secs_link = int($secs / $try); - print STDERR sprintf ("$blurb %-12s %4s (%d/%d);\t %2d secs/link\n", + my $secs_link = $secs / $try; + print STDERR sprintf ("$blurb %-14s %4s (%d/%d);" . + " \t %.1f secs/link\n", "$name:", "$pct%", $suc, $try, $secs_link); } } @@ -2302,7 +2698,7 @@ sub save_recent_url($$) { return unless ($verbose_warnings); $_ = $url; - my ($site) = m@^http://([^ \t\n\r/:]+)@; + my ($site) = m@^https?://([^ \t\n\r/:]+)@; return unless defined ($site); if ($base eq $driftnet_magic || $base eq $local_magic) { @@ -2461,8 +2857,9 @@ sub image_size($) { sub which($) { my ($prog) = @_; foreach (split (/:/, $ENV{PATH})) { - if (-x "$_/$prog") { - return $prog; + my $path = "$_/$prog"; + if (-x $path) { + return $path; } } return undef; @@ -2522,15 +2919,17 @@ sub url_only_output() { # ############################################################################## -my $image_ppm = sprintf ("%s/webcollage-%08x", - ($ENV{TMPDIR} ? $ENV{TMPDIR} : "/tmp"), - rand(0xFFFFFFFF)); -my $image_tmp1 = sprintf ("%s/webcollage-1-%08x", - ($ENV{TMPDIR} ? $ENV{TMPDIR} : "/tmp"), - rand(0xFFFFFFFF)); -my $image_tmp2 = sprintf ("%s/webcollage-2-%08x", - ($ENV{TMPDIR} ? $ENV{TMPDIR} : "/tmp"), - rand(0xFFFFFFFF)); +my ($image_ppm, $image_tmp1, $image_tmp2); +{ + my $seed = rand(0xFFFFFFFF); + $image_ppm = sprintf ("%s/webcollage-%08x", + ($ENV{TMPDIR} ? $ENV{TMPDIR} : "/tmp"), + $seed); + $image_tmp1 = $image_ppm . '-1.ppm'; + $image_tmp2 = $image_ppm . '-2.ppm'; + $image_ppm .= '.ppm'; +} + my $filter_cmd = undef; my $post_filter_cmd = undef; @@ -2641,12 +3040,12 @@ sub image_to_pnm($$$) { $body = undef; }; - if (($pid = open(PIPE, "| $cmd2 > $output"))) { + if (($pid = open (my $pipe, "| $cmd2 > $output"))) { $timed_out = 0; alarm $cvt_timeout; - print PIPE $body; + print $pipe $body; $body = undef; - close PIPE; + close $pipe; LOG ($verbose_exec, "awaiting $pid"); waitpid ($pid, 0); @@ -2701,10 +3100,9 @@ sub ppmmake($$$$) { my $pixel = pack('CCC', $r, $g, $b); my $bits = "P6\n$w $h\n255\n" . ($pixel x ($w * $h)); - local *OUT; - open (OUT, ">$outfile") || error ("$outfile: $!"); - print OUT $bits; - close OUT; + open (my $out, '>', $outfile) || error ("$outfile: $!"); + print $out $bits; + close $out; } @@ -2738,13 +3136,24 @@ my $ppm_to_root_window_cmd = undef; sub x_or_pbm_output($) { my ($window_id) = @_; + # Adjust the PATH for OS X 10.10. + # + $_ = $0; + s:/[^/]*$::; + s/([^a-zA-Z0-9._\-+\/])/\\$1/g; + $ENV{PATH} = "$_:$ENV{PATH}"; + # Check for our helper program, to see whether we need to use PPM pipelines. # $_ = "webcollage-helper"; - if (defined ($webcollage_helper) || which ($_)) { - $webcollage_helper = $_ unless (defined($webcollage_helper)); + + if (! defined ($webcollage_helper)) { + $webcollage_helper = which ($_); + } + + if (defined ($webcollage_helper)) { LOG ($verbose_pbm, "found \"$webcollage_helper\""); - $webcollage_helper .= " -v"; + $webcollage_helper = "'$webcollage_helper' -v"; } else { LOG (($verbose_pbm || $verbose_load), "no $_ program"); } @@ -2769,6 +3178,16 @@ sub x_or_pbm_output($) { which ($_) || error "$_ not found on \$PATH."; } + # If we're using webcollage-helper and not a filter, then the tmp files + # are JPEGs, not PPMs. + # + if (defined ($webcollage_helper) && !defined ($filter_cmd)) { + foreach ($image_ppm, $image_tmp1, $image_tmp2) { + s/\.ppm$/.jpg/s; + } + } + + # find a root-window displayer program. # if (!$no_output_p) { @@ -2843,12 +3262,12 @@ sub x_or_pbm_output($) { my ($iw, $ih); my $body = ""; - local *IMG; - open(IMG, "<$bgimage") || error "couldn't open $bgimage: $!"; - my $cmd; - while () { $body .= $_; } - close (IMG); + open (my $imgf, '<:raw', $bgimage) || error "couldn't open $bgimage: $!"; + local $/ = undef; # read entire file + $body = <$imgf>; + close ($imgf); + my $cmd; if ((@_ = gif_size ($body))) { ($iw, $ih) = @_; $cmd = "giftopnm |"; @@ -2876,10 +3295,10 @@ sub x_or_pbm_output($) { "pasting $bgimage (${iw}x$ih) into base image at $x,$y"); $cmd .= "pnmpaste - $x $y $image_ppm > $image_tmp1"; - open (IMG, "| $cmd") || error "running $cmd: $!"; - print IMG $body; + open ($imgf, "| $cmd") || error "running $cmd: $!"; + print $imgf $body; $body = undef; - close (IMG); + close ($imgf); LOG ($verbose_exec, "subproc exited normally."); rename ($image_tmp1, $image_ppm) || error "renaming $image_tmp1 to $image_ppm: $!"; @@ -2937,10 +3356,17 @@ sub paste_image($$$$) { return 0; } - local *OUT; - open (OUT, ">$image_tmp1") || error ("writing $image_tmp1: $!"); - print OUT $body || error ("writing $image_tmp1: $!"); - close OUT || error ("writing $image_tmp1: $!"); + if ($iw <= 0 || $ih <= 0 || $iw > 9999 || $ih > 9999) { + LOG (($verbose_pbm || $verbose_load), + "ludicrous image dimensions: $iw x $ih (" . length($body) . + "): $img"); + $body = undef; + return 0; + } + + open (my $out, '>:raw', $image_tmp1) || error ("writing $image_tmp1: $!"); + (print $out $body) || error ("writing $image_tmp1: $!"); + close ($out) || error ("writing $image_tmp1: $!"); } else { ($iw, $ih) = image_to_pnm ($img, $body, $image_tmp1); @@ -2972,12 +3398,11 @@ sub paste_image($$$$) { rename ($image_tmp2, $image_tmp1); # re-get the width/height in case the filter resized it. - local *IMG; - open(IMG, "<$image_tmp1") || return 0; - $_ = ; - $_ = ; + open (my $imgf, '<:raw', $image_tmp1) || return 0; + $_ = <$imgf>; + $_ = <$imgf>; ($iw, $ih) = m/^(\d+) (\d+)$/; - close (IMG); + close ($imgf); return 0 unless ($iw && $ih); } @@ -3192,7 +3617,7 @@ sub paste_image($$$$) { # the next network retrieval, which is probably a better thing # to do anyway. # - $cmd .= " &"; + $cmd .= " &" unless ($cocoa_p); $rc = nontrapping_system ($cmd); @@ -3244,6 +3669,8 @@ sub update_imagemap($$$$$$$$) { my $imagemap_html = $imagemap_base . ".html"; my $imagemap_jpg = $imagemap_base . ".jpg"; + my $imagemap_jpg2 = $imagemap_jpg; + $imagemap_jpg2 =~ s@^.*/@@gs; if (!defined ($imagemap_html_tmp)) { $imagemap_html_tmp = $imagemap_html . sprintf (".%08x", rand(0xffffffff)); @@ -3254,16 +3681,16 @@ sub update_imagemap($$$$$$$$) { # my $template_html = ''; { - local *IN; - if (open (IN, "<$imagemap_html")) { - while () { $template_html .= $_; } - close IN; + if (open (my $in, '<', $imagemap_html)) { + local $/ = undef; # read entire file + $template_html = <$in>; + close $in; LOG ($verbose_pbm, "read template $imagemap_html"); } if ($template_html =~ m/^\s*$/s) { $template_html = ("\n" . - "\n"); LOG ($verbose_pbm, "created dummy template"); } @@ -3290,7 +3717,7 @@ sub update_imagemap($$$$$$$$) { my $body = $template_html; my $areas = join ("\n\t", @imagemap_areas); my $map = ("\n\t$areas\n"); - my $img = (""); @@ -3307,10 +3734,9 @@ sub update_imagemap($$$$$$$$) { $body =~ s@().*?()@$1$size$2@si; } - local *OUT; - open (OUT, ">$imagemap_html_tmp") || error ("$imagemap_html_tmp: $!"); - print OUT $body || error ("$imagemap_html_tmp: $!"); - close OUT || error ("$imagemap_html_tmp: $!"); + open (my $out, '>', $imagemap_html_tmp) || error ("$imagemap_html_tmp: $!"); + (print $out $body) || error ("$imagemap_html_tmp: $!"); + close ($out) || error ("$imagemap_html_tmp: $!"); LOG ($verbose_pbm, "wrote $imagemap_html_tmp"); } @@ -3325,6 +3751,49 @@ sub update_imagemap($$$$$$$$) { } +# Figure out what the proxy server should be, either from environment +# variables or by parsing the output of the (MacOS) program "scutil", +# which tells us what the system-wide proxy settings are. +# +sub set_proxy() { + + if (! defined($http_proxy)) { + # historical suckage: the environment variable name is lower case. + $http_proxy = $ENV{http_proxy} || $ENV{HTTP_PROXY}; + } + + if (defined ($http_proxy)) { + if ($http_proxy && $http_proxy =~ m@^https?://([^/]*)/?$@ ) { + # historical suckage: allow "http://host:port" as well as "host:port". + $http_proxy = $1; + } + + } else { + my $proxy_data = `scutil --proxy 2>/dev/null`; + my ($server) = ($proxy_data =~ m/\bHTTPProxy\s*:\s*([^\s]+)/s); + my ($port) = ($proxy_data =~ m/\bHTTPPort\s*:\s*([^\s]+)/s); + # Note: this ignores the "ExceptionsList". + if ($server) { + $http_proxy = $server; + $http_proxy .= ":$port" if $port; + } + } + + delete $ENV{http_proxy}; + delete $ENV{HTTP_PROXY}; + delete $ENV{https_proxy}; + delete $ENV{HTTPS_PROXY}; + delete $ENV{PERL_LWP_ENV_PROXY}; + + if ($http_proxy) { + $http_proxy = 'http://' . $http_proxy; + LOG ($verbose_net, "proxy server: $http_proxy"); + } else { + $http_proxy = undef; # for --proxy '' + } +} + + sub init_signals() { $SIG{HUP} = \&signal_cleanup; @@ -3355,78 +3824,73 @@ sub main() { my $root_p = 0; my $window_id = undef; - # historical suckage: the environment variable name is lower case. - $http_proxy = $ENV{http_proxy} || $ENV{HTTP_PROXY}; - - while ($_ = $ARGV[0]) { - shift @ARGV; - if ($_ eq "-display" || - $_ eq "-displ" || - $_ eq "-disp" || - $_ eq "-dis" || - $_ eq "-dpy" || - $_ eq "-d") { + while ($#ARGV >= 0) { + $_ = shift @ARGV; + if (m/^--?d(i(s(p(l(a(y)?)?)?)?)?)?$/s) { $ENV{DISPLAY} = shift @ARGV; - } elsif ($_ eq "-root") { + } elsif (m/^--?root$/s) { $root_p = 1; - } elsif ($_ eq "-window-id" || $_ eq "--window-id") { + } elsif (m/^--?window-id$/s) { $window_id = shift @ARGV; $root_p = 1; - } elsif ($_ eq "-no-output") { + } elsif (m/^--?no-output$/s) { $no_output_p = 1; - } elsif ($_ eq "-urls-only") { + } elsif (m/^--?urls(-only)?$/s) { $urls_only_p = 1; $no_output_p = 1; - } elsif ($_ eq "-cocoa") { + } elsif (m/^--?cocoa$/s) { $cocoa_p = 1; - } elsif ($_ eq "-imagemap") { + } elsif (m/^--?imagemap$/s) { $imagemap_base = shift @ARGV; $no_output_p = 1; - } elsif ($_ eq "-verbose") { + } elsif (m/^--?verbose$/s) { $verbose++; } elsif (m/^-v+$/) { $verbose += length($_)-1; - } elsif ($_ eq "-delay") { + } elsif (m/^--?delay$/s) { $delay = shift @ARGV; - } elsif ($_ eq "-timeout") { + } elsif (m/^--?timeout$/s) { $http_timeout = shift @ARGV; - } elsif ($_ eq "-filter") { + } elsif (m/^--?filter$/s) { $filter_cmd = shift @ARGV; - } elsif ($_ eq "-filter2") { + } elsif (m/^--?filter2$/s) { $post_filter_cmd = shift @ARGV; - } elsif ($_ eq "-background" || $_ eq "-bg") { + } elsif (m/^--?(background|bg)$/s) { $background = shift @ARGV; - } elsif ($_ eq "-size") { + } elsif (m/^--?size$/s) { $_ = shift @ARGV; if (m@^(\d+)x(\d+)$@) { $img_width = $1; $img_height = $2; } else { - error "argument to \"-size\" must be of the form \"640x400\""; + error "argument to \"--size\" must be of the form \"640x400\""; } - } elsif ($_ eq "-proxy" || $_ eq "-http-proxy") { + } elsif (m/^--?(http-)?proxy$/s) { $http_proxy = shift @ARGV; - } elsif ($_ eq "-dictionary" || $_ eq "-dict") { + } elsif (m/^--?dict(ionary)?$/s) { $dict = shift @ARGV; - } elsif ($_ eq "-opacity") { + } elsif (m/^--?opacity$/s) { $opacity = shift @ARGV; error ("opacity must be between 0.0 and 1.0") if ($opacity <= 0 || $opacity > 1); - } elsif ($_ eq "-driftnet" || $_ eq "--driftnet") { + } elsif (m/^--?driftnet$/s) { @search_methods = ( 100, "driftnet", \&pick_from_driftnet ); if (! ($ARGV[0] =~ m/^-/)) { $driftnet_cmd = shift @ARGV; } else { $driftnet_cmd = $default_driftnet_cmd; } - } elsif ($_ eq "-directory" || $_ eq "--directory") { + } elsif (m/^--?dir(ectory)?$/s) { @search_methods = ( 100, "local", \&pick_from_local_dir ); if (! ($ARGV[0] =~ m/^-/)) { $local_dir = shift @ARGV; } else { error ("local directory path must be set") } - } elsif ($_ eq "-debug" || $_ eq "--debug") { + } elsif (m/^--?fps$/s) { + # -fps only works on MacOS, via "webcollage-cocoa.m". + # Ignore it if passed to this script in an X11 context. + } elsif (m/^--?debug$/s) { my $which = shift @ARGV; my @rest = @search_methods; my $ok = 0; @@ -3443,32 +3907,27 @@ sub main() { } error "no such search method as \"$which\"" unless ($ok); LOG (1, "DEBUG: using only \"$which\""); + $report_performance_interval = 30; } else { + print STDERR "unknown option: $_\n\n"; print STDERR "$copyright\nusage: $progname " . - "[-root] [-display dpy] [-verbose] [-debug which]\n" . - "\t\t [-timeout secs] [-delay secs] [-size WxH]\n" . - "\t\t [-no-output] [-urls-only] [-imagemap filename]\n" . - "\t\t [-filter cmd] [-filter2 cmd] [-background color]\n" . - "\t\t [-dictionary dictionary-file] [-http-proxy host[:port]]\n" . - "\t\t [-driftnet [driftnet-program-and-args]]\n" . - "\t\t [-directory local-image-directory]\n" . + "[--root] [--display dpy] [--verbose] [--debug which]\n" . + "\t\t [--timeout secs] [--delay secs] [--size WxH]\n" . + "\t\t [--no-output] [--urls-only] [--imagemap filename]\n" . + "\t\t [--background color] [--opacity f]\n" . + "\t\t [--filter cmd] [--filter2 cmd]\n" . + "\t\t [--dictionary dictionary-file] [--http-proxy host[:port]]\n" . + "\t\t [--driftnet [driftnet-program-and-args]]\n" . + "\t\t [--directory local-image-directory]\n" . "\n"; exit 1; } } - if ($http_proxy && $http_proxy eq "") { - $http_proxy = undef; - } - if ($http_proxy && $http_proxy =~ m@^http://([^/]*)/?$@ ) { - # historical suckage: allow "http://host:port" as well as "host:port". - $http_proxy = $1; - } - if (!$root_p && !$no_output_p && !$cocoa_p) { print STDERR $copyright; - error "the -root argument is mandatory (for now.)"; + error "the --root argument is mandatory (for now.)"; } if (!$no_output_p && !$cocoa_p && !$ENV{DISPLAY}) { @@ -3534,7 +3993,7 @@ sub main() { } if ($imagemap_base && !($img_width && $img_height)) { - error ("-size WxH is required with -imagemap"); + error ("--size WxH is required with --imagemap"); } if (defined ($local_dir)) { @@ -3543,6 +4002,7 @@ sub main() { } init_signals(); + set_proxy(); spawn_driftnet ($driftnet_cmd) if ($driftnet_cmd);