| 1 |
/* font-manager-orthography.c |
| 2 |
* |
| 3 |
* Copyright (C) 2009-2022 Jerry Casiano |
| 4 |
* |
| 5 |
* This program is free software: you can redistribute it and/or modify |
| 6 |
* it under the terms of the GNU General Public License as published by |
| 7 |
* the Free Software Foundation, either version 3 of the License, or |
| 8 |
* (at your option) any later version. |
| 9 |
* |
| 10 |
* This program is distributed in the hope that it will be useful, |
| 11 |
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 |
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 13 |
* GNU General Public License for more details. |
| 14 |
* |
| 15 |
* You should have received a copy of the GNU General Public License |
| 16 |
* along with this program. |
| 17 |
* |
| 18 |
* If not, see <http://www.gnu.org/licenses/gpl-3.0.txt>. |
| 19 |
*/ |
| 20 |
|
| 21 |
#include "font-manager-orthographies.h" |
| 22 |
|
| 23 |
/** |
| 24 |
* SECTION: font-manager-orthographies |
| 25 |
* @short_description: Font language support |
| 26 |
* @title: Orthographies |
| 27 |
* @include: font-manager-orthographies.h |
| 28 |
*/ |
| 29 |
|
| 30 |
#define N_ARABIC G_N_ELEMENTS(ArabicOrthographies) |
| 31 |
#define N_CHINESE G_N_ELEMENTS(ChineseOrthographies) |
| 32 |
#define N_GREEK G_N_ELEMENTS(GreekOrthographies) |
| 33 |
#define N_JAPANESE G_N_ELEMENTS(JapaneseOrthographies) |
| 34 |
#define N_KOREAN G_N_ELEMENTS(KoreanOrthographies) |
| 35 |
#define N_LATIN G_N_ELEMENTS(LatinOrthographies) |
| 36 |
#define N_MISC G_N_ELEMENTS(UncategorizedOrthographies) |
| 37 |
|
| 38 |
#define GET_OBJECT(n) json_node_get_object((JsonNode *) n) |
| 39 |
#define HAS_COVERAGE(n) JSON_NODE_HOLDS_OBJECT((JsonNode *) n) && json_object_has_member(GET_OBJECT(n), "coverage") |
| 40 |
#define GET_COVERAGE(n) HAS_COVERAGE(n) ? json_object_get_double_member(GET_OBJECT(n), "coverage") : 0.0 |
| 41 |
#define LEN_CHARSET(n) json_array_get_length(json_object_get_array_member(GET_OBJECT(n), "filter")) |
| 42 |
|
| 43 |
static GList * |
| 44 |
_hb_set_to_list (const hb_set_t *charset) |
| 45 |
{ |
| 46 |
GList *result = NULL; |
| 47 |
hb_codepoint_t codepoint = HB_SET_VALUE_INVALID; |
| 48 |
while (hb_set_next(charset, &codepoint)) |
| 49 |
if (font_manager_unicode_unichar_isgraph(codepoint)) |
| 50 |
result = g_list_prepend(result, GINT_TO_POINTER(codepoint)); |
| 51 |
return g_list_reverse(result); |
| 52 |
} |
| 53 |
|
| 54 |
static JsonArray * |
| 55 |
_hb_set_to_json_array (const hb_set_t *charset) |
| 56 |
{ |
| 57 |
JsonArray *result = json_array_new(); |
| 58 |
hb_codepoint_t codepoint = HB_SET_VALUE_INVALID; |
| 59 |
while (hb_set_next(charset, &codepoint)) |
| 60 |
if (font_manager_unicode_unichar_isgraph(codepoint)) |
| 61 |
json_array_add_int_element(result, codepoint); |
| 62 |
return result; |
| 63 |
} |
| 64 |
|
| 65 |
static hb_set_t * |
| 66 |
get_charset_from_font_object (JsonObject *font) |
| 67 |
{ |
| 68 |
hb_blob_t *blob = hb_blob_create_from_file(json_object_get_string_member(font, "filepath")); |
| 69 |
hb_face_t *face = hb_face_create(blob, json_object_get_int_member(font, "findex")); |
| 70 |
hb_set_t *charset = hb_set_create(); |
| 71 |
hb_face_collect_unicodes(face, charset); |
| 72 |
hb_blob_destroy(blob); |
| 73 |
hb_face_destroy(face); |
| 74 |
return charset; |
| 75 |
} |
| 76 |
|
| 77 |
static gint |
| 78 |
sort_by_charset_size (gconstpointer a, gconstpointer b) |
| 79 |
{ |
| 80 |
/* Using variables to avoid unused value warning */ |
| 81 |
gint len_a = LEN_CHARSET(a); |
| 82 |
gint len_b = LEN_CHARSET(b); |
| 83 |
return len_a - len_b; |
| 84 |
} |
| 85 |
|
| 86 |
static gint |
| 87 |
sort_by_coverage (gconstpointer a, gconstpointer b) |
| 88 |
{ |
| 89 |
gint order = (int) GET_COVERAGE(a) - GET_COVERAGE(b); |
| 90 |
return order != 0 ? order : sort_by_charset_size(a, b); |
| 91 |
} |
| 92 |
|
| 93 |
static gchar * |
| 94 |
get_sample_from_charlist (GList *charlist) |
| 95 |
{ |
| 96 |
GString *res = g_string_new(NULL); |
| 97 |
guint length = g_list_length(charlist); |
| 98 |
if (length > 0) |
| 99 |
for (int i = 0; i < 24; i++) { |
| 100 |
int rand = g_random_int_range(0, length); |
| 101 |
gunichar ch = GPOINTER_TO_INT(g_list_nth_data(charlist, rand)); |
| 102 |
g_string_append_unichar(res, ch); |
| 103 |
} |
| 104 |
return g_string_free(res, FALSE); |
| 105 |
} |
| 106 |
|
| 107 |
static gchar * |
| 108 |
get_sample_from_charset (hb_set_t *charset) |
| 109 |
{ |
| 110 |
GList *charlist = _hb_set_to_list(charset); |
| 111 |
gchar *res = get_sample_from_charlist(charlist); |
| 112 |
g_list_free(charlist); |
| 113 |
return res; |
| 114 |
} |
| 115 |
|
| 116 |
static JsonObject * |
| 117 |
get_default_orthography (JsonObject *orthography) |
| 118 |
{ |
| 119 |
GList *orthographies = json_object_get_values(orthography); |
| 120 |
JsonObject *res = NULL; |
| 121 |
if (g_list_length(orthographies) > 0) { |
| 122 |
orthographies = g_list_sort(orthographies, sort_by_coverage); |
| 123 |
JsonNode *node = g_list_nth_data(orthographies, 0); |
| 124 |
if (node && json_node_get_node_type(node) == JSON_NODE_OBJECT) |
| 125 |
res = json_node_get_object(node); |
| 126 |
} |
| 127 |
g_list_free(orthographies); |
| 128 |
return res; |
| 129 |
} |
| 130 |
|
| 131 |
static double |
| 132 |
get_coverage_from_charset (JsonObject *results, |
| 133 |
hb_set_t *charset, |
| 134 |
const FontManagerOrthographyData *data) |
| 135 |
{ |
| 136 |
int hits = 0, tries = 0; |
| 137 |
JsonArray *filter = NULL; |
| 138 |
|
| 139 |
/* If it doesn't contain key there's no point in going further */ |
| 140 |
if (!hb_set_has(charset, data->key)) |
| 141 |
return 0; |
| 142 |
|
| 143 |
if (results) |
| 144 |
filter = json_array_new(); |
| 145 |
|
| 146 |
for (int i = 0; data->values[i] != FONT_MANAGER_END_OF_DATA; i++) { |
| 147 |
|
| 148 |
if (data->values[i] == FONT_MANAGER_START_RANGE_PAIR) { |
| 149 |
|
| 150 |
gunichar start = data->values[++i]; |
| 151 |
gunichar end = data->values[++i]; |
| 152 |
|
| 153 |
for (gunichar codepoint = start; codepoint <= end; codepoint++) { |
| 154 |
tries++; |
| 155 |
if (hb_set_has(charset, codepoint)) |
| 156 |
hits++; |
| 157 |
if (results) |
| 158 |
json_array_add_int_element(filter, (int) codepoint); |
| 159 |
} |
| 160 |
|
| 161 |
} else { |
| 162 |
|
| 163 |
tries++; |
| 164 |
if (hb_set_has(charset, data->values[i])) |
| 165 |
hits++; |
| 166 |
if (results) |
| 167 |
json_array_add_int_element(filter, (int) data->values[i]); |
| 168 |
|
| 169 |
} |
| 170 |
|
| 171 |
} |
| 172 |
|
| 173 |
if (results) |
| 174 |
json_object_set_array_member(results, "filter", filter); |
| 175 |
|
| 176 |
return ((double) 100 * hits/tries ); |
| 177 |
} |
| 178 |
|
| 179 |
static gboolean |
| 180 |
check_orthography (JsonObject *results, |
| 181 |
hb_set_t *charset, |
| 182 |
const FontManagerOrthographyData *data) |
| 183 |
{ |
| 184 |
g_autoptr(JsonObject) res = NULL; |
| 185 |
if (results) |
| 186 |
res = json_object_new(); |
| 187 |
double coverage = get_coverage_from_charset(res, charset, data); |
| 188 |
if (coverage == 0) |
| 189 |
return FALSE; |
| 190 |
if (!results) |
| 191 |
return TRUE; |
| 192 |
json_object_set_string_member(res, "name", data->name); |
| 193 |
json_object_set_string_member(res, "native", data->native); |
| 194 |
json_object_set_string_member(res, "sample", data->sample); |
| 195 |
json_object_set_double_member(res, "coverage", coverage); |
| 196 |
json_object_set_object_member(results, data->name, json_object_ref(res)); |
| 197 |
return TRUE; |
| 198 |
} |
| 199 |
|
| 200 |
static gboolean |
| 201 |
charset_contains_sample_string (const hb_set_t *charset, const char *sample) |
| 202 |
{ |
| 203 |
const char *p = sample; |
| 204 |
while (*p) { |
| 205 |
gunichar codepoint = g_utf8_get_char(p); |
| 206 |
if (!hb_set_has(charset, codepoint)) |
| 207 |
return FALSE; |
| 208 |
p = g_utf8_next_char(p); |
| 209 |
} |
| 210 |
return TRUE; |
| 211 |
} |
| 212 |
|
| 213 |
static void |
| 214 |
check_orthographies (JsonObject *results, |
| 215 |
hb_set_t *charset, |
| 216 |
const FontManagerOrthographyData orth[], |
| 217 |
int len) |
| 218 |
{ |
| 219 |
for (int i = 0; i < len; i++) |
| 220 |
check_orthography(results, charset, &orth[i]); |
| 221 |
return; |
| 222 |
} |
| 223 |
|
| 224 |
static gchar * |
| 225 |
get_default_sample_string_for_orthography (JsonObject *orthography) |
| 226 |
{ |
| 227 |
if (json_object_get_size(orthography) > 0) { |
| 228 |
JsonObject *def = get_default_orthography(orthography); |
| 229 |
if (def && json_object_get_double_member(def, "coverage") > 90) { |
| 230 |
const gchar *sample = NULL; |
| 231 |
if (json_object_has_member(orthography, "sample")) |
| 232 |
sample = json_object_get_string_member(orthography, "sample"); |
| 233 |
if (sample != NULL && g_strcmp0(sample, "") != 0) |
| 234 |
return g_strdup(sample); |
| 235 |
} |
| 236 |
} |
| 237 |
|
| 238 |
if (json_object_has_member(orthography, "Basic Latin")) { |
| 239 |
JsonObject *latin = json_object_get_object_member(orthography, "Basic Latin"); |
| 240 |
if (json_object_get_double_member(latin, "coverage") > 90) { |
| 241 |
PangoLanguage *xx = pango_language_from_string("xx"); |
| 242 |
return g_strdup(pango_language_get_sample_string(xx)); |
| 243 |
} |
| 244 |
} |
| 245 |
|
| 246 |
return NULL; |
| 247 |
} |
| 248 |
|
| 249 |
static gchar * |
| 250 |
get_sample_string (JsonObject *orthography, hb_set_t *charset) |
| 251 |
{ |
| 252 |
const char *local_sample = pango_language_get_sample_string(NULL); |
| 253 |
if (charset_contains_sample_string(charset, local_sample)) |
| 254 |
return NULL; |
| 255 |
gchar *sample = get_default_sample_string_for_orthography(orthography); |
| 256 |
return sample ? sample : get_sample_from_charset(charset); |
| 257 |
} |
| 258 |
|
| 259 |
/** |
| 260 |
* font_manager_get_orthography_results: |
| 261 |
* @font: (nullable) (transfer none): #JsonObject |
| 262 |
* |
| 263 |
* The #JsonObject returned will have the following structure: |
| 264 |
* |
| 265 |
*|[ |
| 266 |
* { |
| 267 |
* "Basic Latin": { |
| 268 |
* "filter": [65, 66, ... 122], |
| 269 |
* "name": "Basic Latin", |
| 270 |
* "native": "Basic Latin", |
| 271 |
* "sample": "AaBbCcGgQqRrSsZz", |
| 272 |
* "coverage": 100.0 |
| 273 |
* }, |
| 274 |
* ..., |
| 275 |
* "sample" : null |
| 276 |
* } |
| 277 |
*]| |
| 278 |
* |
| 279 |
* The returned object contains a member for each orthography detected in @font. |
| 280 |
* |
| 281 |
* sample will be set to %NULL if the font supports rendering the sample string returned |
| 282 |
* by #font_manager_get_localized_pangram, otherwise sample will be set to the |
| 283 |
* sample string from the member with the highest coverage, if that should fail then |
| 284 |
* sample will be set to a string randomly generated from the characters available in @font. |
| 285 |
* |
| 286 |
* Returns: (nullable) (transfer full): #JsonObject containing orthography results |
| 287 |
*/ |
| 288 |
JsonObject * |
| 289 |
font_manager_get_orthography_results (JsonObject *font) |
| 290 |
{ |
| 291 |
hb_set_t *charset = NULL; |
| 292 |
JsonObject *results = json_object_new(); |
| 293 |
|
| 294 |
if (font) |
| 295 |
charset = get_charset_from_font_object(font); |
| 296 |
|
| 297 |
if (charset) { |
| 298 |
if (check_orthography(NULL, charset, LatinOrthographies)) |
| 299 |
check_orthographies(results, charset, LatinOrthographies, N_LATIN); |
| 300 |
|
| 301 |
if (check_orthography(NULL, charset, GreekOrthographies)) |
| 302 |
check_orthographies(results, charset, GreekOrthographies, N_GREEK); |
| 303 |
|
| 304 |
if (check_orthography(NULL, charset, ArabicOrthographies)) |
| 305 |
check_orthographies(results, charset, ArabicOrthographies, N_ARABIC); |
| 306 |
|
| 307 |
check_orthographies(results, charset, ChineseOrthographies, N_CHINESE); |
| 308 |
check_orthographies(results, charset, JapaneseOrthographies, N_JAPANESE); |
| 309 |
check_orthographies(results, charset, KoreanOrthographies, N_KOREAN); |
| 310 |
check_orthographies(results, charset, UncategorizedOrthographies, N_MISC); |
| 311 |
} |
| 312 |
|
| 313 |
if (charset && !hb_set_is_empty(charset)) { |
| 314 |
|
| 315 |
if (json_object_get_size(results) == 0) { |
| 316 |
JsonObject *uncategorized = json_object_new(); |
| 317 |
JsonArray *char_array = _hb_set_to_json_array(charset); |
| 318 |
json_object_set_string_member(uncategorized, "name", "Uncategorized"); |
| 319 |
json_object_set_double_member(uncategorized, "coverage", 100); |
| 320 |
json_object_set_array_member(uncategorized, "filter", char_array); |
| 321 |
json_object_set_object_member(results, "Uncategorized", uncategorized); |
| 322 |
} |
| 323 |
|
| 324 |
g_autofree gchar *sample = get_sample_string(results, charset); |
| 325 |
json_object_set_string_member(results, "sample", sample); |
| 326 |
|
| 327 |
} else { |
| 328 |
|
| 329 |
json_object_set_string_member(results, "sample", NULL); |
| 330 |
|
| 331 |
} |
| 332 |
|
| 333 |
if (charset) |
| 334 |
hb_set_destroy(charset); |
| 335 |
|
| 336 |
return results; |
| 337 |
} |
| 338 |
|
| 339 |
/** |
| 340 |
* font_manager_get_sample_string: |
| 341 |
* @font: #JsonObject |
| 342 |
* |
| 343 |
* Returns: (nullable) (transfer full): A newly allocated string that must be freed with #g_free |
| 344 |
* or %NULL if the systems default language is supported |
| 345 |
*/ |
| 346 |
gchar * |
| 347 |
font_manager_get_sample_string (JsonObject *font) |
| 348 |
{ |
| 349 |
const char *local_sample = pango_language_get_sample_string(NULL); |
| 350 |
hb_set_t *charset = get_charset_from_font_object(font); |
| 351 |
if (charset_contains_sample_string(charset, local_sample)) { |
| 352 |
hb_set_destroy(charset); |
| 353 |
return NULL; |
| 354 |
} |
| 355 |
g_autoptr(JsonObject) orthography = font_manager_get_orthography_results(font); |
| 356 |
gchar *sample = get_default_sample_string_for_orthography(orthography); |
| 357 |
if (!sample) |
| 358 |
sample = get_sample_from_charset(charset); |
| 359 |
hb_set_destroy(charset); |
| 360 |
return sample; |
| 361 |
} |
| 362 |
|