src/test/test-utf8.c

   1 /*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
   2
   3 /***
   4   This file is part of systemd.
   5
   6   Copyright 2013 Dave Reisner
   7
   8   systemd is free software; you can redistribute it and/or modify it
   9   under the terms of the GNU Lesser General Public License as published by
  10   the Free Software Foundation; either version 2.1 of the License, or
  11   (at your option) any later version.
  12
  13   systemd is distributed in the hope that it will be useful, but
  14   WITHOUT ANY WARRANTY; without even the implied warranty of
  15   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  16   Lesser General Public License for more details.
  17
  18   You should have received a copy of the GNU Lesser General Public License
  19   along with systemd; If not, see <http://www.gnu.org/licenses/>.
  20 ***/
  21
  22 #include "alloc-util.h"
  23 #include "string-util.h"
  24 #include "utf8.h"
  25 #include "util.h"
  26
  27 static void test_utf8_is_printable(void) {
  28         assert_se(utf8_is_printable("ascii is valid\tunicode", 22));
  29         assert_se(utf8_is_printable("\342\204\242", 3));
  30         assert_se(!utf8_is_printable("\341\204", 2));
  31         assert_se(utf8_is_printable("ąę", 4));
  32 }
  33
  34 static void test_utf8_is_valid(void) {
  35         assert_se(utf8_is_valid("ascii is valid unicode"));
  36         assert_se(utf8_is_valid("\342\204\242"));
  37         assert_se(!utf8_is_valid("\341\204"));
  38 }
  39
  40 static void test_ascii_is_valid(void) {
  41         assert_se(ascii_is_valid("alsdjf\t\vbarr\nba z"));
  42         assert_se(!ascii_is_valid("\342\204\242"));
  43         assert_se(!ascii_is_valid("\341\204"));
  44 }
  45
  46 static void test_utf8_encoded_valid_unichar(void) {
  47         assert_se(utf8_encoded_valid_unichar("\342\204\242") == 3);
  48         assert_se(utf8_encoded_valid_unichar("\302\256") == 2);
  49         assert_se(utf8_encoded_valid_unichar("a") == 1);
  50         assert_se(utf8_encoded_valid_unichar("\341\204") < 0);
  51         assert_se(utf8_encoded_valid_unichar("\341\204\341\204") < 0);
  52 }
  53
  54 static void test_utf8_escaping(void) {
  55         _cleanup_free_ char *p1, *p2, *p3;
  56
  57         p1 = utf8_escape_invalid("goo goo goo");
  58         puts(p1);
  59         assert_se(utf8_is_valid(p1));
  60
  61         p2 = utf8_escape_invalid("\341\204\341\204");
  62         puts(p2);
  63         assert_se(utf8_is_valid(p2));
  64
  65         p3 = utf8_escape_invalid("\341\204");
  66         puts(p3);
  67         assert_se(utf8_is_valid(p3));
  68 }
  69
  70 static void test_utf8_escaping_printable(void) {
  71         _cleanup_free_ char *p1, *p2, *p3, *p4, *p5, *p6;
  72
  73         p1 = utf8_escape_non_printable("goo goo goo");
  74         puts(p1);
  75         assert_se(utf8_is_valid(p1));
  76
  77         p2 = utf8_escape_non_printable("\341\204\341\204");
  78         puts(p2);
  79         assert_se(utf8_is_valid(p2));
  80
  81         p3 = utf8_escape_non_printable("\341\204");
  82         puts(p3);
  83         assert_se(utf8_is_valid(p3));
  84
  85         p4 = utf8_escape_non_printable("ąę\n가너도루\n1234\n\341\204\341\204\n\001 \019\20\a");
  86         puts(p4);
  87         assert_se(utf8_is_valid(p4));
  88
  89         p5 = utf8_escape_non_printable("\001 \019\20\a");
  90         puts(p5);
  91         assert_se(utf8_is_valid(p5));
  92
  93         p6 = utf8_escape_non_printable("\xef\xbf\x30\x13");
  94         puts(p6);
  95         assert_se(utf8_is_valid(p6));
  96 }
  97
  98 static void test_utf16_to_utf8(void) {
  99         char *a = NULL;
 100         const uint16_t utf16[] = { htole16('a'), htole16(0xd800), htole16('b'), htole16(0xdc00), htole16('c'), htole16(0xd801), htole16(0xdc37) };
 101         const char utf8[] = { 'a', 'b', 'c', 0xf0, 0x90, 0x90, 0xb7, 0 };
 102
 103         a = utf16_to_utf8(utf16, 14);
 104         assert_se(a);
 105         assert_se(streq(a, utf8));
 106
 107         free(a);
 108 }
 109
 110 int main(int argc, char *argv[]) {
 111         test_utf8_is_valid();
 112         test_utf8_is_printable();
 113         test_ascii_is_valid();
 114         test_utf8_encoded_valid_unichar();
 115         test_utf8_escaping();
 116         test_utf8_escaping_printable();
 117         test_utf16_to_utf8();
 118
 119         return 0;
 120 }