# Copyright (C) Internet Systems Consortium, Inc. ("ISC") # # This Source Code Form is subject to the terms of the Mozilla Public # License, v. 2.0. If a copy of the MPL was not distributed with this # file, You can obtain one at http://mozilla.org/MPL/2.0/. # # See the COPYRIGHT file distributed with this work for additional # information regarding copyright ownership. SYSTEMTESTTOP=.. . $SYSTEMTESTTOP/conf.sh # Set known locale for the tests if locale -a | grep -qE "^C\\.(UTF-8|utf8)"; then LC_ALL="C.UTF-8" elif locale -a | grep -qE "^en_US\\.(UTF-8|utf8)"; then LC_ALL="en_US.UTF-8" fi export LC_ALL # This set of tests check the behavior of the IDNA options in "dig". # # "dig" supports two IDNA-related options: # # +[no]idnin - Translates a domain name into punycode format before sending # the query to the server. # # Should the input name be a punycode name, "dig +idnin" will also # validate the punycode, rejecting it if it is invalid. # # +[no]idnout - Translates the received punycode domain names into appropriate # unicode characters before displaying. # # The tests run "dig" against an authoritative server configured with a minimal # root zone and nothing else. As a result, all queries will result in an # NXDOMAIN. The server will return the qname sent, which "dig" will display # according to the options selected. This returned string is compared with # the qname originally sent. # # In the comments below, the following nomenclature (taken from RFC 5890) is # used: # # A-label: Label comprising ASCII characters that starts xn-- and whose # characters after the xn-- are a valid output of the Punycode # algorithm. # # Fake A-label: An A-label whose characters after the xn-- are not valid # Punycode output. # # U-label: Unicode (native character) form of a label. # # For the purpose of this test script, U-labels do not include labels that # comprise purely ASCII characters, which are referred to as "ASCII-labels" # here. Valid ASCII-labels comprise letters, digits and hyphens and do not # start with a hyphen. # # References: # 1. http://www.unicode.org/reports/tr46/#Deviations # 2. http://www.unicode.org/reports/tr46/#IDNAComparison # Using dig insecure mode as we are not testing DNSSEC here DIGCMD="$DIG -i -p ${PORT} @10.53.0.1" # Initialize test count and status return n=0 status=0 # Function for extracting the qname from the response # # This is the first field in the line after the line starting # ";; QUESTION SECTION:". # # The string returned includes the trailing period. qname() { awk 'BEGIN { qs = 0; } \ /;; QUESTION SECTION:/ { qs = 1; next; } \ qs == 1 {sub(";", "", $1) ; print $1; exit 0; }' \ $1 } # Function for performing a test where "dig" is expected to succeed. # # $1 - Description of the test # $2 - Dig command additional options # $3 - Name being queried # $4 - The name that is expected to be displayed by "dig". Note that names # displayed by "dig" will always have a trailing period, so this # parameter should have that period as well. idna_test() { n=`expr $n + 1` description=$1 if [ "$2" != "" ]; then description="${description}: $2" fi echo_i "$description ($n)" ret=0 $DIGCMD $2 $3 > dig.out.$n 2>&1 if [ $? -ne 0 ]; then echo_i "failed: dig command returned non-zero status" ret=1 else actual=`qname dig.out.$n` if [ "$4" != "$actual" ]; then echo_i "failed: expected answer $4, actual result $actual" ret=1 fi fi status=`expr $status + $ret` } # Function for performing a test where "dig" is expected to fail # # $1 - Description of the test # $2 - Dig command additional options # $3 - Name being queried idna_fail() { n=`expr $n + 1` description=$1 if [ "$2" != "" ]; then description="${description}: $2" fi echo_i "$description ($n)" ret=0 $DIGCMD $2 $3 > dig.out.$n 2>&1 if [ $? -eq 0 ]; then echo_i "failed: dig command unexpectedly succeeded" ret=1 fi status=`expr $status + $ret` } # Function to check that case is preserved for an all-ASCII label. # # Without IDNA support, case-preservation is the expected behavior. # # With IDNA support... not really. IDNA maps uppercase ASCII characters to # their lower-case equivalent. When IDNA support in "dig" was updated to # non-transitional IDNA 2008, the switch "+idnin" was added and made the default # behaviour. This meant that the command "dig LocalhosT" (no command switches) # sends the qname "localhost", a change in behavior from earlier versions. # # This was felt to be confusing to the significant number of users who are # not interested in IDNA. For this reason, after "dig" passes the input qname # through the IDNA conversion, is does a case-insensitive comparison with the # result. If the two are the same, "dig" can conclude that the qname is # entirely ASCII and is uses the entered string instead of the converted string # as the qname. ascii_case_preservation_test() { text="Checking valid ASCII label" idna_test "$text" "" LocalhosT LocalhosT. idna_test "$text" "+noidnin +noidnout" LocalhosT LocalhosT. idna_test "$text" "+noidnin +idnout" LocalhosT LocalhosT. idna_test "$text" "+idnin +noidnout" LocalhosT LocalhosT. idna_test "$text" "+idnin +idnout" LocalhosT LocalhosT. } # Function to perform the tests if IDNA is enabled. idna_enabled_test() { echo_i "IDNA is enabled, all IDNA tests will be performed" # Check that case is preserved on an ASCII label. ascii_case_preservation_test # Test of a valid U-label # # +noidnin +noidnout: The label is sent as a unicode octet stream and dig # will display the string in the \nnn format. # +noidnin +idnout: As for the previous case. # +idnin +noidnout: The label is converted to the xn-- format. "dig" # displays the returned xn-- text. # +idnin +idnout: The label is converted to the xn-- format. "dig" # converts the returned xn-- string back to the original # unicode text. # # Note that ASCII characters are converted to lower-case. text="Checking valid non-ASCII label" idna_test "$text" "" "München" "münchen." idna_test "$text" "+noidnin +noidnout" "München" "M\195\188nchen." idna_test "$text" "+noidnin +idnout" "München" "M\195\188nchen." idna_test "$text" "+idnin +noidnout" "München" "xn--mnchen-3ya." idna_test "$text" "+idnin +idnout" "München" "münchen." # Tests of transitional processing of a valid U-label # # IDNA2003 introduced national character sets but, unfortunately, didn't # support several characters properly. One of those was the German # character "ß" (the "Eszett" or "sharp s"), which was interpreted as "ss". # So the domain “faß.de” domain (for example) was processed as “fass.de”. # # This was corrected in IDNA2008, although some vendors that adopted this # standard chose to keep the existing IDNA2003 translation for this # character to prevent problems (e.g. people visiting www.faß.example would, # under IDNA2003, go to www.fass.example but under IDNA2008 would end up at # www.fa\195\159.example - a different web site). # # BIND has adopted a hard transition, so this test checks that these # transitional mapping is not used. The tests are essentially the same as # for the valid U-label. text="Checking that non-transitional IDNA processing is used" idna_test "$text" "" "faß.de" "faß.de." idna_test "$text" "+noidnin +noidnout" "faß.de" "fa\195\159.de." idna_test "$text" "+noidnin +idnout" "faß.de" "fa\195\159.de." idna_test "$text" "+idnin +noidnout" "faß.de" "xn--fa-hia.de." idna_test "$text" "+idnin +idnout" "faß.de" "faß.de." # Another problem character. The final character in the first label mapped # onto the Greek sigma character ("σ") in IDNA2003. text="Second check that non-transitional IDNA processing is used" idna_test "$text" "" "βόλος.com" "βόλος.com." idna_test "$text" "+noidnin +noidnout" "βόλος.com" "\206\178\207\140\206\187\206\191\207\130.com." idna_test "$text" "+noidnin +idnout" "βόλος.com" "\206\178\207\140\206\187\206\191\207\130.com." idna_test "$text" "+idnin +noidnout" "βόλος.com" "xn--nxasmm1c.com." idna_test "$text" "+idnin +idnout" "βόλος.com" "βόλος.com." # Tests of a valid A-label (i.e. starting xn--) # # +noidnout: The string is sent as-is to the server and the returned qname # is displayed in the same form. # +idnout: The string is sent as-is to the server and the returned qname # is displayed as the corresponding U-label. # # The "+[no]idnin" flag has no effect in these cases. text="Checking valid A-label" idna_test "$text" "" "xn--nxasmq6b.com" "βόλοσ.com." idna_test "$text" "+noidnin +noidnout" "xn--nxasmq6b.com" "xn--nxasmq6b.com." idna_test "$text" "+noidnin +idnout" "xn--nxasmq6b.com" "βόλοσ.com." idna_test "$text" "+idnin +noidnout" "xn--nxasmq6b.com" "xn--nxasmq6b.com." idna_test "$text" "+idnin +idnout" "xn--nxasmq6b.com" "βόλοσ.com." # Tests of invalid A-labels # # +noidnin: The label is sent as-is to the server and dig will display the # returned fake A-label in the same form. # +idnin: "dig" should report that the label is not correct. # # +[no]idnout: If the label makes it to the server (via +noidnin), "dig" # should report an error if +idnout is specified. # The minimum length of a punycode A-label is 7 characters. Check that # a shorter label is detected and rejected. text="Checking punycode label shorter than minimum valid length" idna_fail "$text" "" "xn--xx" idna_test "$text" "+noidnin +noidnout" "xn--xx" "xn--xx." idna_fail "$text" "+noidnin +idnout" "xn--xx" idna_fail "$text" "+idnin +noidnout" "xn--xx" idna_fail "$text" "+idnin +idnout" "xn--xx" # Fake A-label - the string does not translate to anything. text="Checking fake A-label" idna_fail "$text" "" "xn--ahahah" idna_test "$text" "+noidnin +noidnout" "xn--ahahah" "xn--ahahah." idna_fail "$text" "+noidnin +idnout" "xn--ahahah" idna_fail "$text" "+idnin +noidnout" "xn--ahahah" idna_fail "$text" "+idnin +idnout" "xn--ahahah" # Too long a label. The punycode string is too long (at 64 characters). # BIND rejects such labels: with +idnin label="xn--xflod18hstflod18hstflod18hstflod18hstflod18hstflod18-1iejjjj" text="Checking punycode label longer than maximum valid length" idna_fail "$text" "" "$label" idna_fail "$text" "+noidnin +noidnout" "$label" idna_fail "$text" "+noidnin +idnout" "$label" idna_fail "$text" "+idnin +noidnout" "$label" idna_fail "$text" "+idnin +idnout" "$label" # Tests of a valid unicode string but an invalid U-label (input) # # Symbols are not valid IDNA names. # # +noidnin: "dig" should send unicode octets to the server and display the # returned qname in the same form. # +idnin: "dig" should generate an error. # # The +[no]idnout options should not have any effect on the test. text="Checking invalid input U-label" idna_fail "$text" "" "🧦.com" idna_test "$text" "+noidnin +noidnout" "🧦.com" "\240\159\167\166.com." idna_test "$text" "+noidnin +idnout" "🧦.com" "\240\159\167\166.com." idna_fail "$text" "+idnin +noidnout" "🧦.com" idna_fail "$text" "+idnin +idnout" "🧦.com" # Tests of a valid unicode string but an invalid U-label (output) # # Symbols are not valid IDNA names. # # Note that an invalid U-label is accepted even when +idnin is in effect # because "xn--19g" is valid Punycode. # # +noidnout: "dig" should send the ACE string to the server and display the # returned qname. # +idnout: "dig" should generate an error. # # The +[no]idnin options should not have any effect on the test. text="Checking invalid output U-label" idna_fail "$text" "" "xn--19g" idna_test "$text" "+noidnin +noidnout" "xn--19g" "xn--19g." idna_fail "$text" "+noidnin +idnout" "xn--19g" idna_test "$text" "+idnin +noidnout" "xn--19g" "xn--19g." idna_fail "$text" "+idnin +idnout" "xn--19g" } # Function to perform tests if IDNA is not enabled. idna_disabled_test() { echo_i "IDNA is disabled, only case mapping tests will be performed" ascii_case_preservation_test } # Main test begins here $FEATURETEST --with-idn if [ $? -eq 0 ]; then idna_enabled_test else idna_disabled_test fi exit $status