src/test/test-utf8.c

   1 /***
   2   This file is part of systemd.
   3
   4   Copyright 2013 Dave Reisner
   5
   6   systemd is free software; you can redistribute it and/or modify it
   7   under the terms of the GNU Lesser General Public License as published by
   8   the Free Software Foundation; either version 2.1 of the License, or
   9   (at your option) any later version.
  10
  11   systemd is distributed in the hope that it will be useful, but
  12   WITHOUT ANY WARRANTY; without even the implied warranty of
  13   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  14   Lesser General Public License for more details.
  15
  16   You should have received a copy of the GNU Lesser General Public License
  17   along with systemd; If not, see <http://www.gnu.org/licenses/>.
  18 ***/
  19
  20 #include "alloc-util.h"
  21 #include "string-util.h"
  22 #include "utf8.h"
  23 #include "util.h"
  24
  25 static void test_utf8_is_printable(void) {
  26         assert_se(utf8_is_printable("ascii is valid\tunicode", 22));
  27         assert_se(utf8_is_printable("\342\204\242", 3));
  28         assert_se(!utf8_is_printable("\341\204", 2));
  29         assert_se(utf8_is_printable("ąę", 4));
  30 }
  31
  32 static void test_utf8_is_valid(void) {
  33         assert_se(utf8_is_valid("ascii is valid unicode"));
  34         assert_se(utf8_is_valid("\342\204\242"));
  35         assert_se(!utf8_is_valid("\341\204"));
  36 }
  37
  38 static void test_ascii_is_valid(void) {
  39         assert_se(ascii_is_valid("alsdjf\t\vbarr\nba z"));
  40         assert_se(!ascii_is_valid("\342\204\242"));
  41         assert_se(!ascii_is_valid("\341\204"));
  42 }
  43
  44 static void test_utf8_encoded_valid_unichar(void) {
  45         assert_se(utf8_encoded_valid_unichar("\342\204\242") == 3);
  46         assert_se(utf8_encoded_valid_unichar("\302\256") == 2);
  47         assert_se(utf8_encoded_valid_unichar("a") == 1);
  48         assert_se(utf8_encoded_valid_unichar("\341\204") < 0);
  49         assert_se(utf8_encoded_valid_unichar("\341\204\341\204") < 0);
  50 }
  51
  52 static void test_utf8_escaping(void) {
  53         _cleanup_free_ char *p1, *p2, *p3;
  54
  55         p1 = utf8_escape_invalid("goo goo goo");
  56         puts(p1);
  57         assert_se(utf8_is_valid(p1));
  58
  59         p2 = utf8_escape_invalid("\341\204\341\204");
  60         puts(p2);
  61         assert_se(utf8_is_valid(p2));
  62
  63         p3 = utf8_escape_invalid("\341\204");
  64         puts(p3);
  65         assert_se(utf8_is_valid(p3));
  66 }
  67
  68 static void test_utf8_escaping_printable(void) {
  69         _cleanup_free_ char *p1, *p2, *p3, *p4, *p5, *p6;
  70
  71         p1 = utf8_escape_non_printable("goo goo goo");
  72         puts(p1);
  73         assert_se(utf8_is_valid(p1));
  74
  75         p2 = utf8_escape_non_printable("\341\204\341\204");
  76         puts(p2);
  77         assert_se(utf8_is_valid(p2));
  78
  79         p3 = utf8_escape_non_printable("\341\204");
  80         puts(p3);
  81         assert_se(utf8_is_valid(p3));
  82
  83         p4 = utf8_escape_non_printable("ąę\n가너도루\n1234\n\341\204\341\204\n\001 \019\20\a");
  84         puts(p4);
  85         assert_se(utf8_is_valid(p4));
  86
  87         p5 = utf8_escape_non_printable("\001 \019\20\a");
  88         puts(p5);
  89         assert_se(utf8_is_valid(p5));
  90
  91         p6 = utf8_escape_non_printable("\xef\xbf\x30\x13");
  92         puts(p6);
  93         assert_se(utf8_is_valid(p6));
  94 }
  95
  96 static void test_utf16_to_utf8(void) {
  97         char *a = NULL;
  98         const uint16_t utf16[] = { htole16('a'), htole16(0xd800), htole16('b'), htole16(0xdc00), htole16('c'), htole16(0xd801), htole16(0xdc37) };
  99         const char utf8[] = { 'a', 'b', 'c', 0xf0, 0x90, 0x90, 0xb7, 0 };
 100
 101         a = utf16_to_utf8(utf16, 14);
 102         assert_se(a);
 103         assert_se(streq(a, utf8));
 104
 105         free(a);
 106 }
 107
 108 int main(int argc, char *argv[]) {
 109         test_utf8_is_valid();
 110         test_utf8_is_printable();
 111         test_ascii_is_valid();
 112         test_utf8_encoded_valid_unichar();
 113         test_utf8_escaping();
 114         test_utf8_escaping_printable();
 115         test_utf16_to_utf8();
 116
 117         return 0;
 118 }