You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

test_parse.c 22 kB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554
  1. #include <assert.h>
  2. #include <stddef.h>
  3. #include <stdio.h>
  4. #include <stdlib.h>
  5. #include <string.h>
  6. #include "json.h"
  7. #include "json_tokener.h"
  8. #include "json_visit.h"
  9. static void test_basic_parse(void);
  10. static void test_utf8_parse(void);
  11. static void test_verbose_parse(void);
  12. static void test_incremental_parse(void);
  13. int main(void)
  14. {
  15. MC_SET_DEBUG(1);
  16. static const char separator[] = "==================================";
  17. test_basic_parse();
  18. puts(separator);
  19. test_utf8_parse();
  20. puts(separator);
  21. test_verbose_parse();
  22. puts(separator);
  23. test_incremental_parse();
  24. puts(separator);
  25. return 0;
  26. }
  27. static json_c_visit_userfunc clear_serializer;
  28. static void do_clear_serializer(json_object *jso);
  29. static void single_basic_parse(const char *test_string, int clear_serializer)
  30. {
  31. json_object *new_obj;
  32. new_obj = json_tokener_parse(test_string);
  33. if (clear_serializer)
  34. do_clear_serializer(new_obj);
  35. printf("new_obj.to_string(%s)=%s\n", test_string, json_object_to_json_string(new_obj));
  36. json_object_put(new_obj);
  37. }
  38. static void test_basic_parse()
  39. {
  40. single_basic_parse("\"\003\"", 0);
  41. single_basic_parse("/* hello */\"foo\"", 0);
  42. single_basic_parse("// hello\n\"foo\"", 0);
  43. single_basic_parse("\"foo\"blue", 0);
  44. single_basic_parse("\'foo\'", 0);
  45. single_basic_parse("\"\\u0041\\u0042\\u0043\"", 0);
  46. single_basic_parse("\"\\u4e16\\u754c\\u00df\"", 0);
  47. single_basic_parse("\"\\u4E16\"", 0);
  48. single_basic_parse("\"\\u4e1\"", 0);
  49. single_basic_parse("\"\\u4e1@\"", 0);
  50. single_basic_parse("\"\\ud840\\u4e16\"", 0);
  51. single_basic_parse("\"\\ud840\"", 0);
  52. single_basic_parse("\"\\udd27\"", 0);
  53. // Test with a "short" high surrogate
  54. single_basic_parse("[9,'\\uDAD", 0);
  55. single_basic_parse("null", 0);
  56. single_basic_parse("NaN", 0);
  57. single_basic_parse("-NaN", 0); /* non-sensical, returns null */
  58. single_basic_parse("Inf", 0); /* must use full string, returns null */
  59. single_basic_parse("inf", 0); /* must use full string, returns null */
  60. single_basic_parse("Infinity", 0);
  61. single_basic_parse("infinity", 0);
  62. single_basic_parse("-Infinity", 0);
  63. single_basic_parse("-infinity", 0);
  64. single_basic_parse("{ \"min\": Infinity, \"max\": -Infinity}", 0);
  65. single_basic_parse("Infinity!", 0);
  66. single_basic_parse("Infinitynull", 0);
  67. single_basic_parse("InfinityXXXX", 0);
  68. single_basic_parse("-Infinitynull", 0);
  69. single_basic_parse("-InfinityXXXX", 0);
  70. single_basic_parse("Infinoodle", 0);
  71. single_basic_parse("InfinAAA", 0);
  72. single_basic_parse("-Infinoodle", 0);
  73. single_basic_parse("-InfinAAA", 0);
  74. single_basic_parse("True", 0);
  75. single_basic_parse("False", 0);
  76. /* not case sensitive */
  77. single_basic_parse("tRue", 0);
  78. single_basic_parse("fAlse", 0);
  79. single_basic_parse("nAn", 0);
  80. single_basic_parse("iNfinity", 0);
  81. single_basic_parse("12", 0);
  82. single_basic_parse("12.3", 0);
  83. single_basic_parse("12.3.4", 0); /* non-sensical, returns null */
  84. /* was returning (int)2015 before patch, should return null */
  85. single_basic_parse("2015-01-15", 0);
  86. /* ...but this works. It's rather inconsistent, and a future major release
  87. * should change the behavior so it either always returns null when extra
  88. * bytes are present (preferred), or always return object created from as much
  89. * as was able to be parsed.
  90. */
  91. single_basic_parse("12.3xxx", 0);
  92. single_basic_parse("{\"FoO\" : -12.3E512}", 0);
  93. single_basic_parse("{\"FoO\" : -12.3e512}", 0);
  94. single_basic_parse("{\"FoO\" : -12.3E51.2}", 0); /* non-sensical, returns null */
  95. single_basic_parse("{\"FoO\" : -12.3E512E12}", 0); /* non-sensical, returns null */
  96. single_basic_parse("[\"\\n\"]", 0);
  97. single_basic_parse("[\"\\nabc\\n\"]", 0);
  98. single_basic_parse("[null]", 0);
  99. single_basic_parse("[]", 0);
  100. single_basic_parse("[false]", 0);
  101. single_basic_parse("[\"abc\",null,\"def\",12]", 0);
  102. single_basic_parse("{}", 0);
  103. single_basic_parse("{ \"foo\": \"bar\" }", 0);
  104. single_basic_parse("{ \'foo\': \'bar\' }", 0);
  105. single_basic_parse("{ \"foo\": \"bar\", \"baz\": null, \"bool0\": true }", 0);
  106. single_basic_parse("{ \"foo\": [null, \"foo\"] }", 0);
  107. single_basic_parse("{ \"abc\": 12, \"foo\": \"bar\", \"bool0\": false, \"bool1\": true, "
  108. "\"arr\": [ 1, 2, 3, null, 5 ] }",
  109. 0);
  110. single_basic_parse("{ \"abc\": \"blue\nred\\ngreen\" }", 0);
  111. // Clear serializer for these tests so we see the actual parsed value.
  112. single_basic_parse("null", 1);
  113. single_basic_parse("false", 1);
  114. single_basic_parse("[0e]", 1);
  115. single_basic_parse("[0e+]", 1);
  116. single_basic_parse("[0e+-1]", 1);
  117. single_basic_parse("\"hello world!\"", 1);
  118. // uint64/int64 range test
  119. single_basic_parse("[9223372036854775806]", 1);
  120. single_basic_parse("[9223372036854775807]", 1);
  121. single_basic_parse("[9223372036854775808]", 1);
  122. single_basic_parse("[-9223372036854775807]", 1);
  123. single_basic_parse("[-9223372036854775808]", 1);
  124. single_basic_parse("[-9223372036854775809]", 1);
  125. single_basic_parse("[18446744073709551614]", 1);
  126. single_basic_parse("[18446744073709551615]", 1);
  127. single_basic_parse("[18446744073709551616]", 1);
  128. }
  129. static void test_utf8_parse()
  130. {
  131. // json_tokener_parse doesn't support checking for byte order marks.
  132. // It's the responsibility of the caller to detect and skip a BOM.
  133. // Both of these checks return null.
  134. char* utf8_bom = "\xEF\xBB\xBF";
  135. char* utf8_bom_and_chars = "\xEF\xBB\xBF{}";
  136. single_basic_parse(utf8_bom, 0);
  137. single_basic_parse(utf8_bom_and_chars, 0);
  138. }
  139. // Clear the re-serialization information that the tokener
  140. // saves to ensure that the output reflects the actual
  141. // values we parsed, rather than just the original input.
  142. static void do_clear_serializer(json_object *jso)
  143. {
  144. json_c_visit(jso, 0, clear_serializer, NULL);
  145. }
  146. static int clear_serializer(json_object *jso, int flags, json_object *parent_jso,
  147. const char *jso_key, size_t *jso_index, void *userarg)
  148. {
  149. if (jso)
  150. json_object_set_serializer(jso, NULL, NULL, NULL);
  151. return JSON_C_VISIT_RETURN_CONTINUE;
  152. }
  153. static void test_verbose_parse()
  154. {
  155. json_object *new_obj;
  156. enum json_tokener_error error = json_tokener_success;
  157. new_obj = json_tokener_parse_verbose("{ foo }", &error);
  158. assert(error == json_tokener_error_parse_object_key_name);
  159. assert(new_obj == NULL);
  160. new_obj = json_tokener_parse("{ foo }");
  161. assert(new_obj == NULL);
  162. new_obj = json_tokener_parse("foo");
  163. assert(new_obj == NULL);
  164. new_obj = json_tokener_parse_verbose("foo", &error);
  165. assert(new_obj == NULL);
  166. /* b/c the string starts with 'f' parsing return a boolean error */
  167. assert(error == json_tokener_error_parse_boolean);
  168. puts("json_tokener_parse_verbose() OK");
  169. }
  170. struct incremental_step
  171. {
  172. const char *string_to_parse;
  173. int length;
  174. int char_offset;
  175. enum json_tokener_error expected_error;
  176. int reset_tokener; /* Set to 1 to call json_tokener_reset() after parsing */
  177. int tok_flags; /* JSON_TOKENER_* flags to pass to json_tokener_set_flags() */
  178. } incremental_steps[] = {
  179. /* Check that full json messages can be parsed, both w/ and w/o a reset */
  180. {"{ \"foo\": 123 }", -1, -1, json_tokener_success, 0},
  181. {"{ \"foo\": 456 }", -1, -1, json_tokener_success, 1},
  182. {"{ \"foo\": 789 }", -1, -1, json_tokener_success, 1},
  183. /* Check the comment parse*/
  184. {"/* hello */{ \"foo\"", -1, -1, json_tokener_continue, 0},
  185. {"/* hello */:/* hello */", -1, -1, json_tokener_continue, 0},
  186. {"\"bar\"/* hello */", -1, -1, json_tokener_continue, 0},
  187. {"}/* hello */", -1, -1, json_tokener_success, 1},
  188. {"/ hello ", -1, 1, json_tokener_error_parse_comment, 1},
  189. {"/* hello\"foo\"", -1, -1, json_tokener_continue, 1},
  190. {"/* hello*\"foo\"", -1, -1, json_tokener_continue, 1},
  191. {"// hello\"foo\"", -1, -1, json_tokener_continue, 1},
  192. /* Check a basic incremental parse */
  193. {"{ \"foo", -1, -1, json_tokener_continue, 0},
  194. {"\": {\"bar", -1, -1, json_tokener_continue, 0},
  195. {"\":13}}", -1, -1, json_tokener_success, 1},
  196. /* Check the UTF-16 surrogate pair */
  197. /* parse one char at every time */
  198. {"\"\\", -1, -1, json_tokener_continue, 0},
  199. {"u", -1, -1, json_tokener_continue, 0},
  200. {"d", -1, -1, json_tokener_continue, 0},
  201. {"8", -1, -1, json_tokener_continue, 0},
  202. {"3", -1, -1, json_tokener_continue, 0},
  203. {"4", -1, -1, json_tokener_continue, 0},
  204. {"\\", -1, -1, json_tokener_continue, 0},
  205. {"u", -1, -1, json_tokener_continue, 0},
  206. {"d", -1, -1, json_tokener_continue, 0},
  207. {"d", -1, -1, json_tokener_continue, 0},
  208. {"1", -1, -1, json_tokener_continue, 0},
  209. {"e\"", -1, -1, json_tokener_success, 1},
  210. /* parse two char at every time */
  211. {"\"\\u", -1, -1, json_tokener_continue, 0},
  212. {"d8", -1, -1, json_tokener_continue, 0},
  213. {"34", -1, -1, json_tokener_continue, 0},
  214. {"\\u", -1, -1, json_tokener_continue, 0},
  215. {"dd", -1, -1, json_tokener_continue, 0},
  216. {"1e\"", -1, -1, json_tokener_success, 1},
  217. /* check the low surrogate pair */
  218. {"\"\\ud834", -1, -1, json_tokener_continue, 0},
  219. {"\\udd1e\"", -1, -1, json_tokener_success, 1},
  220. {"\"\\ud834\\", -1, -1, json_tokener_continue, 0},
  221. {"udd1e\"", -1, -1, json_tokener_success, 1},
  222. {"\"\\ud834\\u", -1, -1, json_tokener_continue, 0},
  223. {"dd1e\"", -1, -1, json_tokener_success, 1},
  224. /* Check that json_tokener_reset actually resets */
  225. {"{ \"foo", -1, -1, json_tokener_continue, 1},
  226. {": \"bar\"}", -1, 0, json_tokener_error_parse_unexpected, 1},
  227. /* Check incremental parsing with trailing characters */
  228. {"{ \"foo", -1, -1, json_tokener_continue, 0},
  229. {"\": {\"bar", -1, -1, json_tokener_continue, 0},
  230. {"\":13}}XXXX", 10, 6, json_tokener_success, 0},
  231. {"XXXX", 4, 0, json_tokener_error_parse_unexpected, 1},
  232. /* Check that trailing characters can change w/o a reset */
  233. {"{\"x\": 123 }\"X\"", -1, 11, json_tokener_success, 0},
  234. {"\"Y\"", -1, -1, json_tokener_success, 1},
  235. /* Trailing characters should cause a failure in strict mode */
  236. {"{\"foo\":9}{\"bar\":8}", -1, 9, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  237. /* ... unless explicitly allowed. */
  238. {"{\"foo\":9}{\"bar\":8}", -1, 9, json_tokener_success, 0,
  239. JSON_TOKENER_STRICT | JSON_TOKENER_ALLOW_TRAILING_CHARS},
  240. {"{\"b\":8}ignored garbage", -1, 7, json_tokener_success, 1,
  241. JSON_TOKENER_STRICT | JSON_TOKENER_ALLOW_TRAILING_CHARS},
  242. /* To stop parsing a number we need to reach a non-digit, e.g. a \0 */
  243. {"1", 1, 1, json_tokener_continue, 0},
  244. /* This should parse as the number 12, since it continues the "1" */
  245. {"2", 2, 1, json_tokener_success, 0},
  246. {"12{", 3, 2, json_tokener_success, 1},
  247. /* Parse number in strict model */
  248. {"[02]", -1, 3, json_tokener_error_parse_number, 1, JSON_TOKENER_STRICT},
  249. /* Similar tests for other kinds of objects: */
  250. /* These could all return success immediately, since regardless of
  251. what follows the false/true/null token we *will* return a json object,
  252. but it currently doesn't work that way. hmm... */
  253. {"false", 5, 5, json_tokener_continue, 1},
  254. {"false", 6, 5, json_tokener_success, 1},
  255. {"true", 4, 4, json_tokener_continue, 1},
  256. {"true", 5, 4, json_tokener_success, 1},
  257. {"null", 4, 4, json_tokener_continue, 1},
  258. {"null", 5, 4, json_tokener_success, 1},
  259. {"Infinity", 9, 8, json_tokener_success, 1},
  260. {"infinity", 9, 8, json_tokener_success, 1},
  261. {"-infinity", 10, 9, json_tokener_success, 1},
  262. {"infinity", 9, 0, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  263. {"-infinity", 10, 1, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  264. {"inf", 3, 3, json_tokener_continue, 0},
  265. {"inity", 6, 5, json_tokener_success, 1},
  266. {"-inf", 4, 4, json_tokener_continue, 0},
  267. {"inity", 6, 5, json_tokener_success, 1},
  268. {"i", 1, 1, json_tokener_continue, 0},
  269. {"n", 1, 1, json_tokener_continue, 0},
  270. {"f", 1, 1, json_tokener_continue, 0},
  271. {"i", 1, 1, json_tokener_continue, 0},
  272. {"n", 1, 1, json_tokener_continue, 0},
  273. {"i", 1, 1, json_tokener_continue, 0},
  274. {"t", 1, 1, json_tokener_continue, 0},
  275. {"y", 1, 1, json_tokener_continue, 0},
  276. {"", 1, 0, json_tokener_success, 1},
  277. {"-", 1, 1, json_tokener_continue, 0},
  278. {"inf", 3, 3, json_tokener_continue, 0},
  279. {"ini", 3, 3, json_tokener_continue, 0},
  280. {"ty", 3, 2, json_tokener_success, 1},
  281. {"-", 1, 1, json_tokener_continue, 0},
  282. {"i", 1, 1, json_tokener_continue, 0},
  283. {"nfini", 5, 5, json_tokener_continue, 0},
  284. {"ty", 3, 2, json_tokener_success, 1},
  285. {"-i", 2, 2, json_tokener_continue, 0},
  286. {"nfinity", 8, 7, json_tokener_success, 1},
  287. {"InfinityX", 10, 8, json_tokener_success, 0},
  288. {"X", 1, 0, json_tokener_error_parse_unexpected, 1},
  289. {"Infinity1234", 13, 8, json_tokener_success, 0},
  290. {"1234", 5, 4, json_tokener_success, 1},
  291. {"Infinity9999", 8, 8, json_tokener_continue, 0},
  292. /* returns the Infinity loaded up by the previous call: */
  293. {"1234", 5, 0, json_tokener_success, 0},
  294. {"1234", 5, 4, json_tokener_success, 1},
  295. /* offset=1 because "n" is the start of "null". hmm... */
  296. {"noodle", 7, 1, json_tokener_error_parse_null, 1},
  297. /* offset=2 because "na" is the start of "nan". hmm... */
  298. {"naodle", 7, 2, json_tokener_error_parse_null, 1},
  299. /* offset=2 because "tr" is the start of "true". hmm... */
  300. {"track", 6, 2, json_tokener_error_parse_boolean, 1},
  301. {"fail", 5, 2, json_tokener_error_parse_boolean, 1},
  302. /* Although they may initially look like they should fail,
  303. * the next few tests check that parsing multiple sequential
  304. * json objects in the input works as expected
  305. */
  306. {"null123", 9, 4, json_tokener_success, 0},
  307. {&"null123"[4], 4, 3, json_tokener_success, 1},
  308. {"nullx", 5, 4, json_tokener_success, 0},
  309. {&"nullx"[4], 2, 0, json_tokener_error_parse_unexpected, 1},
  310. {"{\"a\":1}{\"b\":2}", 15, 7, json_tokener_success, 0},
  311. {&"{\"a\":1}{\"b\":2}"[7], 8, 7, json_tokener_success, 1},
  312. /* Some bad formatting. Check we get the correct error status */
  313. {"2015-01-15", 10, 4, json_tokener_error_parse_number, 1},
  314. /* Strings have a well defined end point, so we can stop at the quote */
  315. {"\"blue\"", -1, -1, json_tokener_success, 0},
  316. /* Check each of the escape sequences defined by the spec */
  317. {"\"\\\"\"", -1, -1, json_tokener_success, 0},
  318. {"\"\\\\\"", -1, -1, json_tokener_success, 0},
  319. {"\"\\b\"", -1, -1, json_tokener_success, 0},
  320. {"\"\\f\"", -1, -1, json_tokener_success, 0},
  321. {"\"\\n\"", -1, -1, json_tokener_success, 0},
  322. {"\"\\r\"", -1, -1, json_tokener_success, 0},
  323. {"\"\\t\"", -1, -1, json_tokener_success, 0},
  324. {"\"\\/\"", -1, -1, json_tokener_success, 0},
  325. // Escaping a forward slash is optional
  326. {"\"/\"", -1, -1, json_tokener_success, 0},
  327. /* Check wrong escape sequences */
  328. {"\"\\a\"", -1, 2, json_tokener_error_parse_string, 1},
  329. /* Check '\'' in strict model */
  330. {"\'foo\'", -1, 0, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  331. /* Parse array/object */
  332. {"[1,2,3]", -1, -1, json_tokener_success, 0},
  333. {"[1,2,3}", -1, 6, json_tokener_error_parse_array, 1},
  334. {"{\"a\"}", -1, 4, json_tokener_error_parse_object_key_sep, 1},
  335. {"{\"a\":1]", -1, 6, json_tokener_error_parse_object_value_sep, 1},
  336. {"{\"a\"::1}", -1, 5, json_tokener_error_parse_unexpected, 1},
  337. {"{\"a\":}", -1, 5, json_tokener_error_parse_unexpected, 1},
  338. {"{\"a\":1,\"a\":2}", -1, -1, json_tokener_success, 1},
  339. {"\"a\":1}", -1, 3, json_tokener_success, 1},
  340. {"{\"a\":1", -1, -1, json_tokener_continue, 1},
  341. {"[,]", -1, 1, json_tokener_error_parse_unexpected, 1},
  342. {"[,1]", -1, 1, json_tokener_error_parse_unexpected, 1},
  343. /* This behaviour doesn't entirely follow the json spec, but until we have
  344. * a way to specify how strict to be we follow Postel's Law and be liberal
  345. * in what we accept (up to a point).
  346. */
  347. {"[1,2,3,]", -1, -1, json_tokener_success, 0},
  348. {"[1,2,,3,]", -1, 5, json_tokener_error_parse_unexpected, 0},
  349. {"[1,2,3,]", -1, 7, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  350. {"{\"a\":1,}", -1, 7, json_tokener_error_parse_unexpected, 1, JSON_TOKENER_STRICT},
  351. // utf-8 test
  352. // acsll encoding
  353. {"\x22\x31\x32\x33\x61\x73\x63\x24\x25\x26\x22", -1, -1, json_tokener_success, 1,
  354. JSON_TOKENER_VALIDATE_UTF8},
  355. {"\x22\x31\x32\x33\x61\x73\x63\x24\x25\x26\x22", -1, -1, json_tokener_success, 1},
  356. // utf-8 encoding
  357. {"\x22\xe4\xb8\x96\xe7\x95\x8c\x22", -1, -1, json_tokener_success, 1,
  358. JSON_TOKENER_VALIDATE_UTF8},
  359. {"\x22\xe4\xb8", -1, 3, json_tokener_error_parse_utf8_string, 0, JSON_TOKENER_VALIDATE_UTF8},
  360. {"\x96\xe7\x95\x8c\x22", -1, 0, json_tokener_error_parse_utf8_string, 1,
  361. JSON_TOKENER_VALIDATE_UTF8},
  362. {"\x22\xe4\xb8\x96\xe7\x95\x8c\x22", -1, -1, json_tokener_success, 1},
  363. {"\x22\xcf\x80\xcf\x86\x22", -1, -1, json_tokener_success, 1, JSON_TOKENER_VALIDATE_UTF8},
  364. {"\x22\xf0\xa5\x91\x95\x22", -1, -1, json_tokener_success, 1, JSON_TOKENER_VALIDATE_UTF8},
  365. // wrong utf-8 encoding
  366. {"\x22\xe6\x9d\x4e\x22", -1, 3, json_tokener_error_parse_utf8_string, 1,
  367. JSON_TOKENER_VALIDATE_UTF8},
  368. {"\x22\xe6\x9d\x4e\x22", -1, 5, json_tokener_success, 1},
  369. // GBK encoding
  370. {"\x22\xc0\xee\xc5\xf4\x22", -1, 2, json_tokener_error_parse_utf8_string, 1,
  371. JSON_TOKENER_VALIDATE_UTF8},
  372. {"\x22\xc0\xee\xc5\xf4\x22", -1, 6, json_tokener_success, 1},
  373. // char after space
  374. {"\x20\x20\x22\xe4\xb8\x96\x22", -1, -1, json_tokener_success, 1, JSON_TOKENER_VALIDATE_UTF8},
  375. {"\x20\x20\x81\x22\xe4\xb8\x96\x22", -1, 2, json_tokener_error_parse_utf8_string, 1,
  376. JSON_TOKENER_VALIDATE_UTF8},
  377. {"\x5b\x20\x81\x31\x5d", -1, 2, json_tokener_error_parse_utf8_string, 1,
  378. JSON_TOKENER_VALIDATE_UTF8},
  379. // char in state inf
  380. {"\x49\x6e\x66\x69\x6e\x69\x74\x79", 9, 8, json_tokener_success, 1},
  381. {"\x49\x6e\x66\x81\x6e\x69\x74\x79", -1, 3, json_tokener_error_parse_utf8_string, 1,
  382. JSON_TOKENER_VALIDATE_UTF8},
  383. // char in escape unicode
  384. {"\x22\x5c\x75\x64\x38\x35\x35\x5c\x75\x64\x63\x35\x35\x22", 15, 14, json_tokener_success, 1,
  385. JSON_TOKENER_VALIDATE_UTF8},
  386. {"\x22\x5c\x75\x64\x38\x35\x35\xc0\x75\x64\x63\x35\x35\x22", -1, 8,
  387. json_tokener_error_parse_utf8_string, 1, JSON_TOKENER_VALIDATE_UTF8},
  388. {"\x22\x5c\x75\x64\x30\x30\x33\x31\xc0\x22", -1, 9, json_tokener_error_parse_utf8_string, 1,
  389. JSON_TOKENER_VALIDATE_UTF8},
  390. // char in number
  391. {"\x31\x31\x81\x31\x31", -1, 2, json_tokener_error_parse_utf8_string, 1,
  392. JSON_TOKENER_VALIDATE_UTF8},
  393. // char in object
  394. {"\x7b\x22\x31\x81\x22\x3a\x31\x7d", -1, 3, json_tokener_error_parse_utf8_string, 1,
  395. JSON_TOKENER_VALIDATE_UTF8},
  396. {NULL, -1, -1, json_tokener_success, 0},
  397. };
  398. static void test_incremental_parse()
  399. {
  400. json_object *new_obj;
  401. enum json_tokener_error jerr;
  402. struct json_tokener *tok;
  403. const char *string_to_parse;
  404. int ii;
  405. int num_ok, num_error;
  406. num_ok = 0;
  407. num_error = 0;
  408. printf("Starting incremental tests.\n");
  409. printf("Note: quotes and backslashes seen in the output here are literal values passed\n");
  410. printf(" to the parse functions. e.g. this is 4 characters: \"\\f\"\n");
  411. string_to_parse = "{ \"foo"; /* } */
  412. printf("json_tokener_parse(%s) ... ", string_to_parse);
  413. new_obj = json_tokener_parse(string_to_parse);
  414. if (new_obj == NULL)
  415. puts("got error as expected");
  416. /* test incremental parsing in various forms */
  417. tok = json_tokener_new();
  418. for (ii = 0; incremental_steps[ii].string_to_parse != NULL; ii++)
  419. {
  420. int this_step_ok = 0;
  421. struct incremental_step *step = &incremental_steps[ii];
  422. int length = step->length;
  423. size_t expected_char_offset;
  424. json_tokener_set_flags(tok, step->tok_flags);
  425. if (length == -1)
  426. length = (int)strlen(step->string_to_parse);
  427. if (step->char_offset == -1)
  428. expected_char_offset = length;
  429. else
  430. expected_char_offset = step->char_offset;
  431. printf("json_tokener_parse_ex(tok, %-12s, %3d) ... ", step->string_to_parse,
  432. length);
  433. new_obj = json_tokener_parse_ex(tok, step->string_to_parse, length);
  434. jerr = json_tokener_get_error(tok);
  435. if (step->expected_error != json_tokener_success)
  436. {
  437. if (new_obj != NULL)
  438. printf("ERROR: invalid object returned: %s\n",
  439. json_object_to_json_string(new_obj));
  440. else if (jerr != step->expected_error)
  441. printf("ERROR: got wrong error: %s\n",
  442. json_tokener_error_desc(jerr));
  443. else if (json_tokener_get_parse_end(tok) != expected_char_offset)
  444. printf("ERROR: wrong char_offset %zu != expected %zu\n",
  445. json_tokener_get_parse_end(tok), expected_char_offset);
  446. else
  447. {
  448. printf("OK: got correct error: %s\n",
  449. json_tokener_error_desc(jerr));
  450. this_step_ok = 1;
  451. }
  452. }
  453. else
  454. {
  455. if (new_obj == NULL &&
  456. !(step->length >= 4 && strncmp(step->string_to_parse, "null", 4) == 0))
  457. printf("ERROR: expected valid object, instead: %s\n",
  458. json_tokener_error_desc(jerr));
  459. else if (json_tokener_get_parse_end(tok) != expected_char_offset)
  460. printf("ERROR: wrong char_offset %zu != expected %zu\n",
  461. json_tokener_get_parse_end(tok), expected_char_offset);
  462. else
  463. {
  464. printf("OK: got object of type [%s]: %s\n",
  465. json_type_to_name(json_object_get_type(new_obj)),
  466. json_object_to_json_string(new_obj));
  467. this_step_ok = 1;
  468. }
  469. }
  470. if (new_obj)
  471. json_object_put(new_obj);
  472. if (step->reset_tokener & 1)
  473. json_tokener_reset(tok);
  474. if (this_step_ok)
  475. num_ok++;
  476. else
  477. num_error++;
  478. }
  479. json_tokener_free(tok);
  480. printf("End Incremental Tests OK=%d ERROR=%d\n", num_ok, num_error);
  481. }