1 /*
2 * Copyright (c) 1998, 2018, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 #if defined(__linux__) || defined(_ALLBSD_SOURCE)
27 #include <stdio.h>
28 #include <ctype.h>
29 #endif
30 #include <pwd.h>
31 #include <locale.h>
32 #ifndef ARCHPROPNAME
33 #error "The macro ARCHPROPNAME has not been defined"
34 #endif
35 #include <sys/utsname.h> /* For os_name and os_version */
36 #include <langinfo.h> /* For nl_langinfo */
37 #include <stdlib.h>
38 #include <string.h>
39 #include <sys/types.h>
40 #include <unistd.h>
41 #include <sys/param.h>
42 #include <time.h>
43 #include <errno.h>
44
45 #ifdef MACOSX
46 #include "java_props_macosx.h"
47 #endif
48
49 #if defined(_ALLBSD_SOURCE)
50 #if !defined(P_tmpdir)
51 #include <paths.h>
52 #define P_tmpdir _PATH_VARTMP
53 #endif
54 #endif
55
56 #include "locale_str.h"
57 #include "java_props.h"
58
59 #if !defined(_ALLBSD_SOURCE)
60 #ifdef __linux__
61 #ifndef CODESET
62 #define CODESET _NL_CTYPE_CODESET_NAME
63 #endif
64 #else
65 #ifdef ALT_CODESET_KEY
66 #define CODESET ALT_CODESET_KEY
67 #endif
68 #endif
69 #endif /* !_ALLBSD_SOURCE */
70
71 /* Take an array of string pairs (map of key->value) and a string (key).
72 * Examine each pair in the map to see if the first string (key) matches the
73 * string. If so, store the second string of the pair (value) in the value and
74 * return 1. Otherwise do nothing and return 0. The end of the map is
75 * indicated by an empty string at the start of a pair (key of "").
76 */
77 static int
78 mapLookup(char* map[], const char* key, char** value) {
79 int i;
80 for (i = 0; strcmp(map[i], ""); i += 2){
81 if (!strcmp(key, map[i])){
82 *value = map[i + 1];
83 return 1;
84 }
85 }
86 return 0;
87 }
88
89 #ifndef P_tmpdir
90 #define P_tmpdir "/var/tmp"
91 #endif
92
93 static int ParseLocale(JNIEnv* env, int cat, char ** std_language, char ** std_script,
94 char ** std_country, char ** std_variant, char ** std_encoding) {
95 char *temp = NULL;
96 char *language = NULL, *country = NULL, *variant = NULL,
97 *encoding = NULL;
98 char *p, *encoding_variant, *old_temp, *old_ev;
99 char *lc;
100
101 /* Query the locale set for the category */
102
103 #ifdef MACOSX
104 lc = setupMacOSXLocale(cat); // malloc'd memory, need to free
105 #else
106 lc = setlocale(cat, NULL);
107 #endif
108
109 #ifndef __linux__
110 if (lc == NULL) {
111 return 0;
112 }
113
114 temp = malloc(strlen(lc) + 1);
115 if (temp == NULL) {
116 #ifdef MACOSX
117 free(lc); // malloced memory
118 #endif
119 JNU_ThrowOutOfMemoryError(env, NULL);
120 return 0;
121 }
122
123 if (cat == LC_CTYPE) {
124 /*
125 * Workaround for Solaris bug 4201684: Xlib doesn't like @euro
126 * locales. Since we don't depend on the libc @euro behavior,
127 * we just remove the qualifier.
128 * On Linux, the bug doesn't occur; on the other hand, @euro
129 * is needed there because it's a shortcut that also determines
130 * the encoding - without it, we wouldn't get ISO-8859-15.
131 * Therefore, this code section is Solaris-specific.
132 */
133 strcpy(temp, lc);
134 p = strstr(temp, "@euro");
135 if (p != NULL) {
136 *p = '\0';
137 setlocale(LC_ALL, temp);
138 }
139 }
140 #else
141 if (lc == NULL || !strcmp(lc, "C") || !strcmp(lc, "POSIX")) {
142 lc = "en_US";
143 }
144
145 temp = malloc(strlen(lc) + 1);
146 if (temp == NULL) {
147 JNU_ThrowOutOfMemoryError(env, NULL);
148 return 0;
149 }
150
151 #endif
152
153 /*
154 * locale string format in Solaris is
155 * <language name>_<country name>.<encoding name>@<variant name>
156 * <country name>, <encoding name>, and <variant name> are optional.
157 */
158
159 strcpy(temp, lc);
160 #ifdef MACOSX
161 free(lc); // malloced memory
162 #endif
163 /* Parse the language, country, encoding, and variant from the
164 * locale. Any of the elements may be missing, but they must occur
165 * in the order language_country.encoding@variant, and must be
166 * preceded by their delimiter (except for language).
167 *
168 * If the locale name (without .encoding@variant, if any) matches
169 * any of the names in the locale_aliases list, map it to the
170 * corresponding full locale name. Most of the entries in the
171 * locale_aliases list are locales that include a language name but
172 * no country name, and this facility is used to map each language
173 * to a default country if that's possible. It's also used to map
174 * the Solaris locale aliases to their proper Java locale IDs.
175 */
176
177 encoding_variant = malloc(strlen(temp)+1);
178 if (encoding_variant == NULL) {
179 free(temp);
180 JNU_ThrowOutOfMemoryError(env, NULL);
181 return 0;
182 }
183
184 if ((p = strchr(temp, '.')) != NULL) {
185 strcpy(encoding_variant, p); /* Copy the leading '.' */
186 *p = '\0';
187 } else if ((p = strchr(temp, '@')) != NULL) {
188 strcpy(encoding_variant, p); /* Copy the leading '@' */
189 *p = '\0';
190 } else {
191 *encoding_variant = '\0';
192 }
193
194 if (mapLookup(locale_aliases, temp, &p)) {
195 old_temp = temp;
196 temp = realloc(temp, strlen(p)+1);
197 if (temp == NULL) {
198 free(old_temp);
199 free(encoding_variant);
200 JNU_ThrowOutOfMemoryError(env, NULL);
201 return 0;
202 }
203 strcpy(temp, p);
204 old_ev = encoding_variant;
205 encoding_variant = realloc(encoding_variant, strlen(temp)+1);
206 if (encoding_variant == NULL) {
207 free(old_ev);
208 free(temp);
209 JNU_ThrowOutOfMemoryError(env, NULL);
210 return 0;
211 }
212 // check the "encoding_variant" again, if any.
213 if ((p = strchr(temp, '.')) != NULL) {
214 strcpy(encoding_variant, p); /* Copy the leading '.' */
215 *p = '\0';
216 } else if ((p = strchr(temp, '@')) != NULL) {
217 strcpy(encoding_variant, p); /* Copy the leading '@' */
218 *p = '\0';
219 }
220 }
221
222 language = temp;
223 if ((country = strchr(temp, '_')) != NULL) {
224 *country++ = '\0';
225 }
226
227 p = encoding_variant;
228 if ((encoding = strchr(p, '.')) != NULL) {
229 p[encoding++ - p] = '\0';
230 p = encoding;
231 }
232 if ((variant = strchr(p, '@')) != NULL) {
233 p[variant++ - p] = '\0';
234 }
235
236 /* Normalize the language name */
237 if (std_language != NULL) {
238 *std_language = "en";
239 if (language != NULL && mapLookup(language_names, language, std_language) == 0) {
240 *std_language = malloc(strlen(language)+1);
241 strcpy(*std_language, language);
242 }
243 }
244
245 /* Normalize the country name */
246 if (std_country != NULL && country != NULL) {
247 if (mapLookup(country_names, country, std_country) == 0) {
248 *std_country = malloc(strlen(country)+1);
249 strcpy(*std_country, country);
250 }
251 }
252
253 /* Normalize the script and variant name. Note that we only use
254 * variants listed in the mapping array; others are ignored.
255 */
256 if (variant != NULL) {
257 if (std_script != NULL) {
258 mapLookup(script_names, variant, std_script);
259 }
260
261 if (std_variant != NULL) {
262 mapLookup(variant_names, variant, std_variant);
263 }
264 }
265
266 /* Normalize the encoding name. Note that we IGNORE the string
267 * 'encoding' extracted from the locale name above. Instead, we use the
268 * more reliable method of calling nl_langinfo(CODESET). This function
269 * returns an empty string if no encoding is set for the given locale
270 * (e.g., the C or POSIX locales); we use the default ISO 8859-1
271 * converter for such locales.
272 */
273 if (std_encoding != NULL) {
274 /* OK, not so reliable - nl_langinfo() gives wrong answers on
275 * Euro locales, in particular. */
276 if (strcmp(p, "ISO8859-15") == 0)
277 p = "ISO8859-15";
278 else
279 p = nl_langinfo(CODESET);
280
281 /* Convert the bare "646" used on Solaris to a proper IANA name */
282 if (strcmp(p, "646") == 0)
283 p = "ISO646-US";
284
285 /* return same result nl_langinfo would return for en_UK,
286 * in order to use optimizations. */
287 *std_encoding = (*p != '\0') ? p : "ISO8859-1";
288
289 #ifdef __linux__
290 /*
291 * Remap the encoding string to a different value for japanese
292 * locales on linux so that customized converters are used instead
293 * of the default converter for "EUC-JP". The customized converters
294 * omit support for the JIS0212 encoding which is not supported by
295 * the variant of "EUC-JP" encoding used on linux
296 */
297 if (strcmp(p, "EUC-JP") == 0) {
298 *std_encoding = "EUC-JP-LINUX";
299 }
300 #endif
301
302 #ifdef _AIX
303 if (strcmp(p, "big5") == 0) {
304 /* On AIX Traditional Chinese Big5 codeset is mapped to IBM-950 */
305 *std_encoding = "IBM-950";
306 } else if (strcmp(p, "IBM-943") == 0) {
307 /*
308 * On AIX, IBM-943 is mapped to IBM-943C in which symbol 'yen' and
309 * 'overline' are replaced with 'backslash' and 'tilde' from ASCII
310 * making first 96 code points same as ASCII.
311 */
312 *std_encoding = "IBM-943C";
313 }
314 #endif
315
316 #ifdef __solaris__
317 if (strcmp(p,"eucJP") == 0) {
318 /* For Solaris use customized vendor defined character
319 * customized EUC-JP converter
320 */
321 *std_encoding = "eucJP-open";
322 } else if (strcmp(p, "Big5") == 0 || strcmp(p, "BIG5") == 0) {
323 /*
324 * Remap the encoding string to Big5_Solaris which augments
325 * the default converter for Solaris Big5 locales to include
326 * seven additional ideographic characters beyond those included
327 * in the Java "Big5" converter.
328 */
329 *std_encoding = "Big5_Solaris";
330 } else if (strcmp(p, "Big5-HKSCS") == 0) {
331 /*
332 * Solaris uses HKSCS2001
333 */
334 *std_encoding = "Big5-HKSCS-2001";
335 }
336 #endif
337 #ifdef MACOSX
338 /*
339 * For the case on MacOS X where encoding is set to US-ASCII, but we
340 * don't have any encoding hints from LANG/LC_ALL/LC_CTYPE, use UTF-8
341 * instead.
342 *
343 * The contents of ASCII files will still be read and displayed
344 * correctly, but so will files containing UTF-8 characters beyond the
345 * standard ASCII range.
346 *
347 * Specifically, this allows apps launched by double-clicking a .jar
348 * file to correctly read UTF-8 files using the default encoding (see
349 * 8011194).
350 */
351 const char* env_lang = getenv("LANG");
352 const char* env_lc_all = getenv("LC_ALL");
353 const char* env_lc_ctype = getenv("LC_CTYPE");
354
355 if (strcmp(p,"US-ASCII") == 0 &&
356 (env_lang == NULL || strlen(env_lang) == 0) &&
357 (env_lc_all == NULL || strlen(env_lc_all) == 0) &&
358 (env_lc_ctype == NULL || strlen(env_lc_ctype) == 0)) {
359 *std_encoding = "UTF-8";
360 }
361 #endif
362 }
363
364 free(temp);
365 free(encoding_variant);
366
367 return 1;
368 }
369
370 /* This function gets called very early, before VM_CALLS are setup.
371 * Do not use any of the VM_CALLS entries!!!
372 */
373 java_props_t *
374 GetJavaProperties(JNIEnv *env)
375 {
376 static java_props_t sprops;
377 char *v; /* tmp var */
378
379 if (sprops.user_dir) {
380 return &sprops;
381 }
382
383 /* tmp dir */
384 sprops.tmp_dir = P_tmpdir;
385 #ifdef MACOSX
386 /* darwin has a per-user temp dir */
387 static char tmp_path[PATH_MAX];
388 int pathSize = confstr(_CS_DARWIN_USER_TEMP_DIR, tmp_path, PATH_MAX);
389 if (pathSize > 0 && pathSize <= PATH_MAX) {
390 sprops.tmp_dir = tmp_path;
391 }
392 #endif /* MACOSX */
393
394 /* patches/service packs installed */
395 sprops.patch_level = NULL; // leave it undefined
396
397 #ifdef SI_ISALIST
398 /* supported instruction sets */
399 {
400 char list[258];
401 sysinfo(SI_ISALIST, list, sizeof(list));
402 sprops.cpu_isalist = strdup(list);
403 }
404 #else
405 sprops.cpu_isalist = NULL;
406 #endif
407
408 /* endianness of platform */
409 {
410 unsigned int endianTest = 0xff000000;
411 if (((char*)(&endianTest))[0] != 0)
412 sprops.cpu_endian = "big";
413 else
414 sprops.cpu_endian = "little";
415 }
416
417 /* os properties */
418 {
419 #ifdef MACOSX
420 setOSNameAndVersion(&sprops);
421 #else
422 struct utsname name;
423 uname(&name);
424 sprops.os_name = strdup(name.sysname);
425 #ifdef _AIX
426 {
427 char *os_version = malloc(strlen(name.version) +
428 strlen(name.release) + 2);
429 if (os_version != NULL) {
430 strcpy(os_version, name.version);
431 strcat(os_version, ".");
432 strcat(os_version, name.release);
433 }
434 sprops.os_version = os_version;
435 }
436 #else
437 sprops.os_version = strdup(name.release);
438 #endif /* _AIX */
439 #endif /* MACOSX */
440
441 sprops.os_arch = ARCHPROPNAME;
442 }
443
444 /* ABI property (optional) */
445 #ifdef JDK_ARCH_ABI_PROP_NAME
446 sprops.sun_arch_abi = JDK_ARCH_ABI_PROP_NAME;
447 #endif
448
449 /* Determine the language, country, variant, and encoding from the host,
450 * and store these in the user.language, user.country, user.variant and
451 * file.encoding system properties. */
452 setlocale(LC_ALL, "");
453 if (ParseLocale(env, LC_CTYPE,
454 &(sprops.format_language),
455 &(sprops.format_script),
456 &(sprops.format_country),
457 &(sprops.format_variant),
458 &(sprops.encoding))) {
459 ParseLocale(env, LC_MESSAGES,
460 &(sprops.display_language),
461 &(sprops.display_script),
462 &(sprops.display_country),
463 &(sprops.display_variant),
464 NULL);
465 } else {
466 sprops.display_language = "en";
467 sprops.encoding = "ISO8859-1";
468 }
469
470 /* ParseLocale failed with OOME */
471 JNU_CHECK_EXCEPTION_RETURN(env, NULL);
472
473 #ifdef MACOSX
474 sprops.sun_jnu_encoding = "UTF-8";
475 #else
476 sprops.sun_jnu_encoding = sprops.encoding;
477 #endif
478
479 #ifdef _ALLBSD_SOURCE
480 #if BYTE_ORDER == _LITTLE_ENDIAN
481 sprops.unicode_encoding = "UnicodeLittle";
482 #else
483 sprops.unicode_encoding = "UnicodeBig";
484 #endif
485 #else /* !_ALLBSD_SOURCE */
486 #ifdef __linux__
487 #if __BYTE_ORDER == __LITTLE_ENDIAN
488 sprops.unicode_encoding = "UnicodeLittle";
489 #else
490 sprops.unicode_encoding = "UnicodeBig";
491 #endif
492 #else
493 sprops.unicode_encoding = "UnicodeBig";
494 #endif
495 #endif /* _ALLBSD_SOURCE */
496
497 /* user properties */
498 {
499 struct passwd *pwent = getpwuid(getuid());
500 sprops.user_name = pwent ? strdup(pwent->pw_name) : "?";
501 #ifdef MACOSX
502 setUserHome(&sprops);
503 #else
504 sprops.user_home = pwent ? strdup(pwent->pw_dir) : NULL;
505 #endif
506 if (sprops.user_home == NULL) {
507 sprops.user_home = "?";
508 }
509 }
510
511 /* User TIMEZONE
512 * We defer setting up timezone until it's actually necessary.
513 * Refer to TimeZone.getDefault(). The system property
514 * is able to be set by the command line interface -Duser.timezone.
515 */
516 tzset(); /* for compatibility */
517
518 /* Current directory */
519 {
520 char buf[MAXPATHLEN];
521 errno = 0;
522 if (getcwd(buf, sizeof(buf)) == NULL)
523 JNU_ThrowByName(env, "java/lang/Error",
524 "Properties init: Could not determine current working directory.");
525 else
526 sprops.user_dir = strdup(buf);
527 }
528
529 sprops.file_separator = "/";
530 sprops.path_separator = ":";
531 sprops.line_separator = "\n";
532
533 #ifdef MACOSX
534 setProxyProperties(&sprops);
535 #endif
536
537 return &sprops;
538 }
539
540 jstring
541 GetStringPlatform(JNIEnv *env, nchar* cstr)
542 {
543 return JNU_NewStringPlatform(env, cstr);
544 }