1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/os/persistentdata/persistentstorage/sqlite3api/TEST/TclScript/enc.test Fri Jun 15 03:10:57 2012 +0200
1.3 @@ -0,0 +1,172 @@
1.4 +# 2002 May 24
1.5 +#
1.6 +# The author disclaims copyright to this source code. In place of
1.7 +# a legal notice, here is a blessing:
1.8 +#
1.9 +# May you do good and not evil.
1.10 +# May you find forgiveness for yourself and forgive others.
1.11 +# May you share freely, never taking more than you give.
1.12 +#
1.13 +#***********************************************************************
1.14 +# This file implements regression tests for SQLite library. The focus of
1.15 +# this file is testing the SQLite routines used for converting between the
1.16 +# various suported unicode encodings (UTF-8, UTF-16, UTF-16le and
1.17 +# UTF-16be).
1.18 +#
1.19 +# $Id: enc.test,v 1.7 2007/05/23 16:23:09 danielk1977 Exp $
1.20 +
1.21 +set testdir [file dirname $argv0]
1.22 +source $testdir/tester.tcl
1.23 +
1.24 +# Skip this test if the build does not support multiple encodings.
1.25 +#
1.26 +ifcapable {!utf16} {
1.27 + finish_test
1.28 + return
1.29 +}
1.30 +
1.31 +proc do_bincmp_test {testname got expect} {
1.32 + binary scan $expect \c* expectvals
1.33 + binary scan $got \c* gotvals
1.34 + do_test $testname [list set dummy $gotvals] $expectvals
1.35 +}
1.36 +
1.37 +# $utf16 is a UTF-16 encoded string. Swap each pair of bytes around
1.38 +# to change the byte-order of the string.
1.39 +proc swap_byte_order {utf16} {
1.40 + binary scan $utf16 \c* ints
1.41 +
1.42 + foreach {a b} $ints {
1.43 + lappend ints2 $b
1.44 + lappend ints2 $a
1.45 + }
1.46 +
1.47 + return [binary format \c* $ints2]
1.48 +}
1.49 +
1.50 +#
1.51 +# Test that the SQLite routines for converting between UTF encodings
1.52 +# produce the same results as their TCL counterparts.
1.53 +#
1.54 +# $testname is the prefix to be used for the test names.
1.55 +# $str is a string to use for testing (encoded in UTF-8, as normal for TCL).
1.56 +#
1.57 +# The test procedure is:
1.58 +# 1. Convert the string from UTF-8 to UTF-16le and check that the TCL and
1.59 +# SQLite routines produce the same results.
1.60 +#
1.61 +# 2. Convert the string from UTF-8 to UTF-16be and check that the TCL and
1.62 +# SQLite routines produce the same results.
1.63 +#
1.64 +# 3. Use the SQLite routines to convert the native machine order UTF-16
1.65 +# representation back to the original UTF-8. Check that the result
1.66 +# matches the original representation.
1.67 +#
1.68 +# 4. Add a byte-order mark to each of the UTF-16 representations and
1.69 +# check that the SQLite routines can convert them back to UTF-8. For
1.70 +# byte-order mark info, refer to section 3.10 of the unicode standard.
1.71 +#
1.72 +# 5. Take the byte-order marked UTF-16 strings from step 4 and ensure
1.73 +# that SQLite can convert them both to native byte order UTF-16
1.74 +# strings, sans BOM.
1.75 +#
1.76 +# Coverage:
1.77 +#
1.78 +# sqlite_utf8to16be (step 2)
1.79 +# sqlite_utf8to16le (step 1)
1.80 +# sqlite_utf16to8 (steps 3, 4)
1.81 +# sqlite_utf16to16le (step 5)
1.82 +# sqlite_utf16to16be (step 5)
1.83 +#
1.84 +proc test_conversion {testname str} {
1.85 +
1.86 + # Step 1.
1.87 + set utf16le_sqlite3 [test_translate $str UTF8 UTF16LE]
1.88 + set utf16le_tcl [encoding convertto unicode $str]
1.89 + append utf16le_tcl "\x00\x00"
1.90 + if { $::tcl_platform(byteOrder)!="littleEndian" } {
1.91 + set utf16le_tcl [swap_byte_order $utf16le_tcl]
1.92 + }
1.93 + do_bincmp_test $testname.1 $utf16le_sqlite3 $utf16le_tcl
1.94 + set utf16le $utf16le_tcl
1.95 +
1.96 + # Step 2.
1.97 + set utf16be_sqlite3 [test_translate $str UTF8 UTF16BE]
1.98 + set utf16be_tcl [encoding convertto unicode $str]
1.99 + append utf16be_tcl "\x00\x00"
1.100 + if { $::tcl_platform(byteOrder)=="littleEndian" } {
1.101 + set utf16be_tcl [swap_byte_order $utf16be_tcl]
1.102 + }
1.103 + do_bincmp_test $testname.2 $utf16be_sqlite3 $utf16be_tcl
1.104 + set utf16be $utf16be_tcl
1.105 +
1.106 + # Step 3.
1.107 + if { $::tcl_platform(byteOrder)=="littleEndian" } {
1.108 + set utf16 $utf16le
1.109 + } else {
1.110 + set utf16 $utf16be
1.111 + }
1.112 + set utf8_sqlite3 [test_translate $utf16 UTF16 UTF8]
1.113 + do_bincmp_test $testname.3 $utf8_sqlite3 [binarize $str]
1.114 +
1.115 + # Step 4 (little endian).
1.116 + append utf16le_bom "\xFF\xFE" $utf16le
1.117 + set utf8_sqlite3 [test_translate $utf16le_bom UTF16 UTF8 1]
1.118 + do_bincmp_test $testname.4.le $utf8_sqlite3 [binarize $str]
1.119 +
1.120 + # Step 4 (big endian).
1.121 + append utf16be_bom "\xFE\xFF" $utf16be
1.122 + set utf8_sqlite3 [test_translate $utf16be_bom UTF16 UTF8]
1.123 + do_bincmp_test $testname.4.be $utf8_sqlite3 [binarize $str]
1.124 +
1.125 + # Step 5 (little endian to little endian).
1.126 + set utf16_sqlite3 [test_translate $utf16le_bom UTF16LE UTF16LE]
1.127 + do_bincmp_test $testname.5.le.le $utf16_sqlite3 $utf16le
1.128 +
1.129 + # Step 5 (big endian to big endian).
1.130 + set utf16_sqlite3 [test_translate $utf16be_bom UTF16 UTF16BE]
1.131 + do_bincmp_test $testname.5.be.be $utf16_sqlite3 $utf16be
1.132 +
1.133 + # Step 5 (big endian to little endian).
1.134 + set utf16_sqlite3 [test_translate $utf16be_bom UTF16 UTF16LE]
1.135 + do_bincmp_test $testname.5.be.le $utf16_sqlite3 $utf16le
1.136 +
1.137 + # Step 5 (little endian to big endian).
1.138 + set utf16_sqlite3 [test_translate $utf16le_bom UTF16 UTF16BE]
1.139 + do_bincmp_test $testname.5.le.be $utf16_sqlite3 $utf16be
1.140 +}
1.141 +
1.142 +translate_selftest
1.143 +
1.144 +test_conversion enc-1 "hello world"
1.145 +test_conversion enc-2 "sqlite"
1.146 +test_conversion enc-3 ""
1.147 +test_conversion enc-X "\u0100"
1.148 +test_conversion enc-4 "\u1234"
1.149 +test_conversion enc-5 "\u4321abc"
1.150 +test_conversion enc-6 "\u4321\u1234"
1.151 +test_conversion enc-7 [string repeat "abcde\u00EF\u00EE\uFFFCabc" 100]
1.152 +test_conversion enc-8 [string repeat "\u007E\u007F\u0080\u0081" 100]
1.153 +test_conversion enc-9 [string repeat "\u07FE\u07FF\u0800\u0801\uFFF0" 100]
1.154 +test_conversion enc-10 [string repeat "\uE000" 100]
1.155 +
1.156 +proc test_collate {enc zLeft zRight} {
1.157 + return [string compare $zLeft $zRight]
1.158 +}
1.159 +add_test_collate $::DB 0 0 1
1.160 +do_test enc-11.1 {
1.161 + execsql {
1.162 + CREATE TABLE ab(a COLLATE test_collate, b);
1.163 + INSERT INTO ab VALUES(CAST (X'C388' AS TEXT), X'888800');
1.164 + INSERT INTO ab VALUES(CAST (X'C0808080808080808080808080808080808080808080808080808080808080808080808080808080808080808080808080808080808388' AS TEXT), X'888800');
1.165 + CREATE INDEX ab_i ON ab(a, b);
1.166 + }
1.167 +} {}
1.168 +do_test enc-11.2 {
1.169 + set cp200 "\u00C8"
1.170 + execsql {
1.171 + SELECT count(*) FROM ab WHERE a = $::cp200;
1.172 + }
1.173 +} {2}
1.174 +
1.175 +finish_test