Use whois.nic.$TLD for the "new" gTLDs

The "new" gTLDs are contractually required by ICANN to provide port 43
whois service on this standard domain.

7 files changed, 324 insertions, 0 deletions

diff --git a/.gitignore b/.gitignore
index f990909..dabbc22 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,4 +1,5 @@
 *_del.h
+new_gtlds.h
 tld_serv.h
 servers_charset.h
 version.h
diff --git a/Makefile b/Makefile
index d71d81d..8f1ca35 100644
--- a/Makefile
+++ b/Makefile
@@ -84,6 +84,9 @@ ip_del.h: ip_del_list make_ip_del.pl
 ip6_del.h: ip6_del_list make_ip6_del.pl
 	$(PERL) make_ip6_del.pl < $< > $@
 
+new_gtlds.h: new_gtlds_list make_new_gtlds.pl
+	$(PERL) make_new_gtlds.pl < $< > $@
+
 tld_serv.h: tld_serv_list make_tld_serv.pl
 	$(PERL) make_tld_serv.pl < $< > $@
 
diff --git a/data.h b/data.h
index d25c6dc..96178d2 100644
--- a/data.h
+++ b/data.h
@@ -134,6 +134,11 @@ const struct as32_del as32_assign[] = {
     { 0, 0, NULL }
 };
 
+const char *new_gtlds[] = {
+#include "new_gtlds.h"
+    NULL
+};
+
 const char *tld_serv[] = {
 #include "tld_serv.h"
     NULL,	NULL
diff --git a/make_new_gtlds.pl b/make_new_gtlds.pl
new file mode 100755
index 0000000..964a2f7
--- /dev/null
+++ b/make_new_gtlds.pl
@@ -0,0 +1,16 @@
+#!/usr/bin/perl
+
+use warnings;
+use strict;
+
+while (<>) {
+	chomp;
+	s/#.*$//;
+	s/^\s+//; s/\s+$//;
+	next if /^$/;
+
+	die "format error: $_" if not /^((?:xn--)?[a-z0-9]+)$/;
+
+	print qq|    ".$_",\n|;
+}
+
diff --git a/new_gtlds_list b/new_gtlds_list
new file mode 100644
index 0000000..d545419
--- /dev/null
+++ b/new_gtlds_list
@@ -0,0 +1,266 @@
+# If a TLD is listed in this file then queries will go to whois.nic.$TLD.
+# All "new" gTLDs are mandated by the ICANN contract to provide port 43 and
+# web-based whois service on this standard domain.
+# Any exceptions can be handled in tld_serv_list as usual, since it will
+# be checked first.
+
+academy
+actor
+agency
+airforce
+archi
+associates
+axa
+bar
+bargains
+berlin
+best
+bid
+bike
+black
+blackfriday
+blue
+boutique
+build
+builders
+buzz
+cab
+camera
+camp
+capital
+cards
+care
+career
+careers
+cash
+catering
+center
+ceo
+cheap
+christmas
+citic
+cleaning
+clinic
+clothing
+club
+codes
+coffee
+college
+cologne
+com
+community
+company
+computer
+condos
+construction
+consulting
+contractors
+cooking
+cool
+country
+creditcard
+cruises
+dance
+dating
+democrat
+dental
+desi
+diamonds
+directory
+discount
+dnp
+domains
+education
+email
+engineering
+enterprises
+equipment
+estate
+eus
+events
+exchange
+expert
+exposed
+fail
+farm
+feedback
+finance
+financial
+fish
+fishing
+fitness
+flights
+florist
+foo
+foundation
+frogans
+fund
+furniture
+futbol
+gal
+gallery
+gift
+glass
+gop
+graphics
+gratis
+gripe
+guitars
+guru
+haus
+holdings
+holiday
+horse
+house
+immobilien
+industries
+info
+ink
+institute
+insure
+international
+investments
+jetzt
+kaufen
+kim
+kitchen
+kiwi
+koeln
+kred
+land
+lease
+lighting
+limited
+limo
+link
+london
+luxury
+maison
+management
+mango
+marketing
+media
+meet
+menu
+miami
+moda
+moe
+monash
+moscow
+nagoya
+net
+neustar
+ninja
+nyc
+okinawa
+onl
+org
+paris
+partners
+parts
+photo
+photography
+photos
+pics
+pictures
+pink
+plumbing
+productions
+properties
+pub
+qpon
+quebec
+recipes
+red
+reisen
+ren
+rentals
+repair
+report
+rest
+reviews
+rich
+rocks
+rodeo
+ruhr
+ryukyu
+saarland
+schule
+services
+sexy
+shiksha
+shoes
+singles
+social
+sohu
+solar
+solutions
+soy
+supplies
+supply
+support
+surgery
+systems
+tattoo
+tax
+technology
+tienda
+tips
+today
+tokyo
+tools
+town
+toys
+trade
+training
+university
+uno
+vacations
+vegas
+ventures
+viajes
+villas
+vision
+vodka
+vote
+voting
+voto
+voyage
+wang
+watch
+webcam
+wed
+wien
+wiki
+works
+wtc
+wtf
+xn--3bst00m
+xn--3ds443g
+xn--55qw42g
+xn--55qx5d
+xn--6frz82g
+xn--6qq986b3xl
+xn--80adxhks
+xn--80asehdb
+xn--80aswg
+xn--c1avg
+xn--cg4bki
+xn--czru2d
+xn--d1acj3b
+xn--fiq228c5hs
+xn--fiq64b
+xn--i1b6b1a6a2e
+xn--io0a7i
+xn--mgbab2bd
+xn--ngbc5azd
+xn--nqv7f
+xn--nqv7fs00ema
+xn--q9jyb4c
+xn--rhqv96g
+xn--ses554g
+xn--unup4y
+xn--zfr164b
+xyz
+yokohama
+zone
diff --git a/whois.c b/whois.c
index e1fb9e3..7476ca2 100644
--- a/whois.c
+++ b/whois.c
@@ -459,6 +459,21 @@ char *guess_server(const char *s)
 	if (domcmp(s, tld_serv[i]))
 	    return strdup(tld_serv[i + 1]);
 
+    /* use the default server name for "new" gTLDs */
+    if (is_new_gtld(s)) {
+	char *server;
+	const char *p, *tld = NULL;
+
+	for (p = s; *p; p++)		/* look for the TLD */
+	    if (*p == '.')
+		tld = p;
+
+	server = malloc(strlen("whois.nic") + strlen(tld) + 1);
+	strcpy(server, "whois.nic");
+	strcat(server, tld);
+	return(server);
+    }
+
     /* no dot but hyphen */
     if (!strchr(s, '.')) {
 	/* search for strings at the start of the word */
@@ -535,6 +550,12 @@ char *queryformat(const char *server, const char *flags, const char *query)
 	    }
 	    break;
 	}
+
+    /* Use UTF-8 by default for "new" gTLDs */
+    if (!simple_recode_input_charset &&		/* was not in the database */
+	    !strchr(query, ' ') &&		/* and has no parameters */
+	    is_new_gtld(query))			/* and is a "new" gTLD: */
+	simple_recode_input_charset = "utf-8";	/* then try UTF-8 */
 #endif
 
 #ifdef HAVE_LIBIDN
@@ -957,6 +978,17 @@ int domcmp(const char *dom, const char *tld)
     return 0;
 }
 
+int is_new_gtld(const char *s)
+{
+    int i;
+
+    for (i = 0; new_gtlds[i]; i++)
+	if (domcmp(s, new_gtlds[i]))
+	    return 1;
+
+    return 0;
+}
+
 /*
  * Attempt to normalize a query by removing trailing dots and whitespace,
  * then convert the domain to punycode.
diff --git a/whois.h b/whois.h
index a9b9a64..cec65b3 100644
--- a/whois.h
+++ b/whois.h
@@ -28,6 +28,7 @@ unsigned long myinet_aton(const char *);
 unsigned long asn32_to_long(const char *);
 int isasciidigit(const char);
 int domcmp(const char *, const char *);
+int is_new_gtld(const char *);
 int domfind(const char *, const char *[]);
 char *normalize_domain(const char *);
 char *convert_6to4(const char *);