1 AT_BANNER([i18n recoding])
3 m4_divert_push([PREPARE_TESTS])
4 supports_encodings () {
6 *-*-linux* | *-*-*-gnu*)
7 dnl GNU/Linux always has the encodings we want. We can't ask
8 dnl config.charset about them because it has a special case here
9 dnl too and won't tell us.
13 for encoding in "$@"; do
14 $SHELL $top_srcdir/gl/config.charset "$host" | grep '$2' || return 77
19 m4_divert_pop([PREPARE_TESTS])
21 # CHECK_I18N_RECODE([TITLE], [FROM-CODING], [TO-CODING],
22 # [FROM-TEXT], [TO-TEXT])
24 # Converts FROM-TEXT from FROM-CODING to TO-CODING and checks that the result
25 # is TO-TEXT. The "printf" program is applied to both FROM-TEXT and TO-TEXT to
26 # allow for backslash-escapes. (Hex escapes are not portable; use octal
28 m4_define([CHECK_I18N_RECODE],
29 [AT_SETUP([convert $1])
32 dnl Skip the test if this host doesn't know the source and target encodings.
33 AT_CHECK([supports_encodings '$2' '$3'])
34 AT_CHECK_UNQUOTED([i18n-test recode '$2' '$3' `printf '$4'`], [0], [`printf '$5'`
38 CHECK_I18N_RECODE([reflexively], [ASCII], [ASCII], [abc], [abc])
39 CHECK_I18N_RECODE([without any change], [ASCII], [UTF-8], [abc], [abc])
41 CHECK_I18N_RECODE([from ISO-8859-1 to UTF-8], [ISO-8859-1], [UTF-8],
43 CHECK_I18N_RECODE([from UTF-8 to ISO-8859-1], [UTF-8], [ISO-8859-1],
46 # 0xc0 == 0300 is invalid in UTF-8
47 CHECK_I18N_RECODE([invalid UTF-8 to ISO-8859-1], [UTF-8], [ISO-8859-1],
49 # 0xc2 == 0302 is the first byte of a 2-byte UTF-8 sequence
50 CHECK_I18N_RECODE([truncated UTF-8 to ISO-8559-1], [UTF-8], [ISO-8859-1],
53 dnl The input to this test is 7 bytes long and the expected output is 9 bytes.
54 dnl So it should exercise the E2BIG case
55 CHECK_I18N_RECODE([from ISO-8859-1 to UTF-8 with overflow],
56 [ISO-8859-1], [UTF-8],
57 [Tsch\374\337!], [Tsch\303\274\303\237!])
59 AT_SETUP([convert unknown encoding])
61 AT_CHECK([i18n-test recode nonexistent1 nonexistent2 asdf], [0], [asdf
63 [Warning: cannot create a converter for `nonexistent1' to `nonexistent2': Invalid argument
67 AT_BANNER([i18n concatenation])
69 # CHECK_I18N_CONCAT([HEAD], [TAIL], [ENCODING], [MAX-LEN], [ANSWER])
71 # Concatenates HEAD and TAIL, omitting as many characters from HEAD as needed
72 # to make the result come out to no more than MAX-LEN bytes if it was expressed
73 # in ENCODING, and checks that the answer matches ANSWER. HEAD, TAIL, and
74 # ANSWER are all in UTF-8. The "printf" program is applied to HEAD, TAIL, and
75 # ANSWER to allow for backslash-escapes. (Hex escapes are not portable; use
76 # octal escapes instead.)
77 m4_define([CHECK_I18N_CONCAT],
78 [AT_SETUP([m4_if([$2], [], [truncate "$1" to $4 bytes in $3],
79 [truncate "$1" + "$2" to $4 bytes in $3])])
82 dnl Skip the test if this host doesn't know the encoding.
83 AT_CHECK([supports_encodings '$3'])
85 [i18n-test concat "`printf '$1'`" "`printf '$2'`" '$3' '$4'], [0],
90 CHECK_I18N_CONCAT([abc], [], [UTF-8], [6], [abc])
91 CHECK_I18N_CONCAT([], [xyz], [UTF-8], [6], [xyz])
92 CHECK_I18N_CONCAT([], [], [UTF-8], [6], [])
93 CHECK_I18N_CONCAT([abcdefghij], [], [UTF-8], [6], [abcdef])
94 CHECK_I18N_CONCAT([], [tuvwxyz], [UTF-8], [6], [tuvwxyz])
96 CHECK_I18N_CONCAT([abc], [xyz], [UTF-8], [6], [abcxyz])
97 CHECK_I18N_CONCAT([abcd], [xyz], [UTF-8], [6], [abcxyz])
98 CHECK_I18N_CONCAT([abc], [uvwxyz], [UTF-8], [6], [uvwxyz])
100 # x in a box ( x⃞ ) is U+0078, U+20DE, 4 bytes in UTF-8, and one grapheme
102 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [0], [y])
103 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [1], [y])
104 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [2], [y])
105 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [3], [y])
106 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [4], [y])
107 CHECK_I18N_CONCAT([x\342\203\236], [y], [UTF-8], [5], [x\342\203\236y])
109 # éèä is only 3 bytes in ISO-8859-1.
110 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [0], [xyz])
111 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [1], [xyz])
112 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [2], [xyz])
113 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [3], [xyz])
114 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [4],
116 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [5],
117 [\303\251\303\250xyz])
118 CHECK_I18N_CONCAT([\303\251\303\250\303\244], [xyz], [ISO-8859-1], [6],
119 [\303\251\303\250\303\244xyz])