| 1 |
/* |
| 2 |
textprint: Test filtering of text before terminal printout. |
| 3 |
|
| 4 |
copyright 2019-2020 by the mpg123 project |
| 5 |
free software under the terms of the LGPL 2.1 |
| 6 |
|
| 7 |
see COPYING and AUTHORS files in distribution or http://mpg123.org |
| 8 |
initially written by Thomas Orgis |
| 9 |
*/ |
| 10 |
|
| 11 |
#include "../mpg123app.h" |
| 12 |
#include "../local.h" |
| 13 |
|
| 14 |
// A number of UTF-8 strings to test. |
| 15 |
|
| 16 |
struct test_string |
| 17 |
{ |
| 18 |
const char* in; // UTF-8 input including control chars etc. |
| 19 |
const char* ascii; // output reduced to ASCII |
| 20 |
const char* utf8hack; // output in UTF-8 without controls, hacked |
| 21 |
const char* utf8full; // output in UTF-8 only including printing chars |
| 22 |
size_t chars; // number of apparent characters (for utf8hack) |
| 23 |
size_t width; // proper printing width |
| 24 |
}; |
| 25 |
|
| 26 |
// const char *bla = "..." is not constant enoug to server as initializer. |
| 27 |
#define simple_string "Just a plain string without trouble." |
| 28 |
#define simple_width 36 |
| 29 |
// o-Umlaut and a random Chinese character |
| 30 |
#define utf_string "St\xc3\xb6rt mein \xe7\x8c\x95?" |
| 31 |
#define utf_string_ascii "St?rt mein ??" |
| 32 |
#define utf_width 14 |
| 33 |
#define utf_chars 13 |
| 34 |
|
| 35 |
const struct test_string test_input[] = |
| 36 |
{ |
| 37 |
{ |
| 38 |
.in=simple_string |
| 39 |
, .ascii=simple_string |
| 40 |
, .utf8hack=simple_string |
| 41 |
, .utf8full=simple_string |
| 42 |
, .chars=simple_width |
| 43 |
, .width=simple_width |
| 44 |
} |
| 45 |
, { |
| 46 |
.in="U\xcc\x88""bel\nist mir." |
| 47 |
, .ascii="U?bel ist mir." |
| 48 |
, .utf8hack="U\xcc\x88""bel ist mir." |
| 49 |
, .utf8full="U\xcc\x88""bel ist mir." |
| 50 |
, .chars=14 |
| 51 |
, .width=13 |
| 52 |
} |
| 53 |
, { |
| 54 |
.in=utf_string |
| 55 |
, .ascii=utf_string_ascii |
| 56 |
, .utf8hack=utf_string |
| 57 |
, .utf8full=utf_string |
| 58 |
, .chars=utf_chars |
| 59 |
, .width=utf_width |
| 60 |
} |
| 61 |
, { |
| 62 |
.in="Gimme\x0a some\x0a\x0d""con\302\237tr\xc3\xb6l!\x7f\x7f""b \033]0;xtermed\x07" |
| 63 |
, .ascii="Gimme some contr?l!b ]0;xtermed" |
| 64 |
, .utf8hack="Gimme some contr\xc3\xb6l!b ]0;xtermed" |
| 65 |
, .utf8full="Gimme some contr\xc3\xb6l!b ]0;xtermed" |
| 66 |
, .chars=33 |
| 67 |
, .width=33 |
| 68 |
} |
| 69 |
}; |
| 70 |
|
| 71 |
|
| 72 |
int main() |
| 73 |
{ |
| 74 |
char *dst = NULL; |
| 75 |
const size_t test_count = sizeof(test_input)/sizeof(*test_input); |
| 76 |
int err = 0; |
| 77 |
|
| 78 |
// First, without any locale check, we should work like in C locale. |
| 79 |
// Only ASCII is safe. |
| 80 |
fprintf(stderr, "Testing plain ASCII world.\n"); |
| 81 |
for(size_t t=0; t<test_count; ++t) |
| 82 |
{ |
| 83 |
int lerr = 0; |
| 84 |
fprintf(stderr, "string %zu: ", t); |
| 85 |
size_t w = utf8outstr(&dst, test_input[t].in, 1); |
| 86 |
size_t l = strlen(test_input[t].ascii); |
| 87 |
if(!dst) |
| 88 |
{ |
| 89 |
++lerr; |
| 90 |
fprintf(stderr, "(conversion failed) "); |
| 91 |
} |
| 92 |
else if(w == l) |
| 93 |
{ |
| 94 |
if(strcmp(test_input[t].ascii, dst)) |
| 95 |
{ |
| 96 |
++lerr; |
| 97 |
fprintf(stderr, "(mismatch) "); |
| 98 |
} |
| 99 |
} |
| 100 |
else |
| 101 |
{ |
| 102 |
++lerr; |
| 103 |
fprintf(stderr, "(bad length %zu != %zu) ", w, l); |
| 104 |
} |
| 105 |
err += lerr; |
| 106 |
fprintf(stderr, "%s\n", lerr ? "FAIL" : "PASS"); |
| 107 |
} |
| 108 |
|
| 109 |
// Now, the internal UTF-8 sanitation is employed, assuming no working |
| 110 |
// locale to help us. |
| 111 |
utf8env = 1; |
| 112 |
utf8loc = 0; |
| 113 |
fprintf(stderr, "Testing hacked UTF-8 world.\n"); |
| 114 |
for(size_t t=0; t<test_count; ++t) |
| 115 |
{ |
| 116 |
int lerr = 0; |
| 117 |
fprintf(stderr, "string %zu: ", t); |
| 118 |
size_t w = utf8outstr(&dst, test_input[t].in, 1); |
| 119 |
size_t l = test_input[t].chars; |
| 120 |
if(!dst) |
| 121 |
{ |
| 122 |
++lerr; |
| 123 |
fprintf(stderr, "(conversion failed) "); |
| 124 |
} |
| 125 |
else if(w == l) |
| 126 |
{ |
| 127 |
if(strcmp(test_input[t].utf8hack, dst)) |
| 128 |
{ |
| 129 |
++lerr; |
| 130 |
fprintf(stderr, "(mismatch) "); |
| 131 |
} |
| 132 |
} |
| 133 |
else |
| 134 |
{ |
| 135 |
++lerr; |
| 136 |
fprintf(stderr, "(bad length %zu != %zu) ", w, l); |
| 137 |
} |
| 138 |
err += lerr; |
| 139 |
fprintf(stderr, "%s\n", lerr ? "FAIL" : "PASS"); |
| 140 |
} |
| 141 |
|
| 142 |
#if defined(HAVE_MBSTOWCS) && defined(HAVE_WCSWIDTH) && \ |
| 143 |
defined(HAVE_ISWPRINT) && defined(HAVE_WCSTOMBS) |
| 144 |
// Finally, proper UTF-8 handling including iswprint() shall be undertaken. |
| 145 |
// The test is skipped if there is no UTF-8 locale to work with. |
| 146 |
utf8env = 0; |
| 147 |
utf8loc = 0; |
| 148 |
utf8force = 1; |
| 149 |
check_locale(); |
| 150 |
if(utf8env && utf8loc) |
| 151 |
{ |
| 152 |
fprintf(stderr, "Got locale, testing proper workings.\n"); |
| 153 |
for(size_t t=0; t<test_count; ++t) |
| 154 |
{ |
| 155 |
int lerr = 0; |
| 156 |
fprintf(stderr, "string %zu: ", t); |
| 157 |
size_t w = utf8outstr(&dst, test_input[t].in, 1); |
| 158 |
size_t l = test_input[t].width; |
| 159 |
if(!dst) |
| 160 |
{ |
| 161 |
++lerr; |
| 162 |
fprintf(stderr, "(conversion failed) "); |
| 163 |
} |
| 164 |
else if(w == l) |
| 165 |
{ |
| 166 |
if(strcmp(test_input[t].utf8full, dst)) |
| 167 |
{ |
| 168 |
++lerr; |
| 169 |
fprintf(stderr, "(mismatch) "); |
| 170 |
} |
| 171 |
} |
| 172 |
else |
| 173 |
{ |
| 174 |
++lerr; |
| 175 |
fprintf(stderr, "(bad length %zu != %zu) ", w, l); |
| 176 |
} |
| 177 |
err += lerr; |
| 178 |
fprintf(stderr, "%s\n", lerr ? "FAIL" : "PASS"); |
| 179 |
fprintf(stderr, "non-terminal non-filtering: "); |
| 180 |
lerr = 0; |
| 181 |
w = utf8outstr(&dst, test_input[t].in, 0); |
| 182 |
if(!dst || strcmp(test_input[t].in, dst)) |
| 183 |
++lerr; |
| 184 |
err += lerr; |
| 185 |
fprintf(stderr, "%s\n", lerr ? "FAIL" : "PASS"); |
| 186 |
} |
| 187 |
} |
| 188 |
else |
| 189 |
#endif |
| 190 |
fprintf(stderr, "Skipped locale-based conversion\n"); |
| 191 |
|
| 192 |
free(dst); |
| 193 |
|
| 194 |
printf("%s\n", err ? "FAIL" : "PASS"); |
| 195 |
return err ? 1 : 0; |
| 196 |
} |