aboutsummaryrefslogtreecommitdiff
path: root/jim-json.c
blob: 2d613b78d7c7c00176001ecd22efed5b5628d4e6 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
/*
 * Copyright (c) 2015 - 2016 Svyatoslav Mishyn <juef@openmailbox.org>
 * Copyright (c) 2019 Steve Bennett <steveb@workware.net.au>
 *
 * Permission to use, copy, modify, and/or distribute this software for
 * any purpose with or without fee is hereby granted, provided that the
 * above copyright notice and this permission notice appear in all
 * copies.
 *
 * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
 * WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
 * WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
 * AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
 * DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
 * PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
 * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
 * PERFORMANCE OF THIS SOFTWARE.
 */

#include <assert.h>
#include <stddef.h>
#include <stdlib.h>
#include <string.h>

#include <jim.h>

#include "jsmn/jsmn.h"

/* These are all the schema types we support */
typedef enum {
	JSON_BOOL,
	JSON_OBJ,
	JSON_LIST,
	JSON_MIXED,
	JSON_STR,
	JSON_NUM,
	JSON_MAX_TYPE,
} json_schema_t;

struct json_state {
	Jim_Obj *nullObj;
	const char *json;
	jsmntok_t *tok;
	int need_subst;
	/* The following are used for -schema */
	int enable_schema;
	int enable_index;
	Jim_Obj *schemaObj;
	Jim_Obj *schemaTypeObj[JSON_MAX_TYPE];
};

static void json_decode_dump_value(Jim_Interp *interp, struct json_state *state, Jim_Obj *list);

/**
 * Start a new subschema. Returns the previous schemaObj.
 * Does nothing and returns NULL if -schema is not enabled.
 */
static Jim_Obj *json_decode_schema_push(Jim_Interp *interp, struct json_state *state)
{
	Jim_Obj *prevSchemaObj = NULL;
	if (state->enable_schema) {
		prevSchemaObj = state->schemaObj;
		state->schemaObj = Jim_NewListObj(interp, NULL, 0);
		Jim_IncrRefCount(state->schemaObj);
	}
	return prevSchemaObj;
}

/**
 * Combines the current schema with the previous schema, prevSchemaObj
 * returned by json_decode_schema_push().
 * Does nothing if -schema is not enabled.
 */
static void json_decode_schema_pop(Jim_Interp *interp, struct json_state *state, Jim_Obj *prevSchemaObj)
{
	if (state->enable_schema) {
		Jim_ListAppendElement(interp, prevSchemaObj, state->schemaObj);
		Jim_DecrRefCount(interp, state->schemaObj);
		state->schemaObj = prevSchemaObj;
	}
}

/**
 * Appends the schema type to state->schemaObj based on 'type'
 */
static void json_decode_add_schema_type(Jim_Interp *interp, struct json_state *state, json_schema_t type)
{
	static const char * const schema_names[] = {
		"bool",
		"obj",
		"list",
		"mixed",
		"str",
		"num",
	};
	assert(type >= 0 && type < JSON_MAX_TYPE);
	/* Share multiple instances of the same type */
	if (state->schemaTypeObj[type] == NULL) {
		state->schemaTypeObj[type] = Jim_NewStringObj(interp, schema_names[type], -1);
	}
	Jim_ListAppendElement(interp, state->schemaObj, state->schemaTypeObj[type]);
}

/**
 * Returns the schema type for the given token.
 * There is a one-to-one correspondence except for JSMN_PRIMITIVE
 * which will return JSON_BOOL for true, false and JSON_NUM otherise.
 */
static json_schema_t json_decode_get_type(const jsmntok_t *tok, const char *json)
{
	switch (tok->type) {
		case JSMN_PRIMITIVE:
			assert(json);
			if (json[tok->start] == 't' || json[tok->start] == 'f') {
				return JSON_BOOL;
			}
			return JSON_NUM;
		case JSMN_OBJECT:
			return JSON_OBJ;
		case JSMN_ARRAY:
			/* Return mixed by default - need other checks to select list instead */
			return JSON_MIXED;
		case JSMN_STRING:
		default:
			return JSON_STR;
	}
}

/**
 * Returns the current object (state->tok) as a Tcl list.
 *
 * state->tok is incremented to just past the object that was dumped.
 */
static Jim_Obj *
json_decode_dump_container(Jim_Interp *interp, struct json_state *state)
{
	int i;
	Jim_Obj *list = Jim_NewListObj(interp, NULL, 0);
	int size = state->tok->size;
	int type = state->tok->type;
	json_schema_t container_type = JSON_OBJ; /* JSON_LIST, JSON_MIXED or JSON_OBJ */

	if (state->schemaObj) {
		/* Don't strictly need to initialise this, but some compilers can't figure out it is always
		 * assigned a value below.
		 */
		json_schema_t list_type = JSON_STR;
		/* Figure out the type to use for the container */
		if (type == JSMN_ARRAY) {
			/* If every element of the array is of the same primitive schema type (str, bool or num),
			 * we can use "list", otherwise need to use "mixed"
			 */
			container_type = JSON_LIST;
			if (size) {
				list_type = json_decode_get_type(&state->tok[1], state->json);

				if (list_type == JSON_BOOL || list_type == JSON_STR || list_type == JSON_NUM) {
					for (i = 2; i <= size; i++) {
						if (json_decode_get_type(state->tok + i, state->json) != list_type) {
							/* Can't use list */
							container_type = JSON_MIXED;
							break;
						}
					}
				}
				else {
					container_type = JSON_MIXED;
				}
			}
		}
		json_decode_add_schema_type(interp, state, container_type);
		if (container_type == JSON_LIST && size) {
			json_decode_add_schema_type(interp, state, list_type);
		}
	}

	state->tok++;

	for (i = 0; i < size; i++) {
		if (type == JSMN_OBJECT) {
			/* Dump the object key */
			if (state->enable_schema) {
				const char *p = state->json + state->tok->start;
				int len = state->tok->end - state->tok->start;
				Jim_ListAppendElement(interp, state->schemaObj, Jim_NewStringObj(interp, p, len));
			}
			json_decode_dump_value(interp, state, list);
		}

		if (state->enable_index && type == JSMN_ARRAY) {
			Jim_ListAppendElement(interp, list, Jim_NewIntObj(interp, i));
		}

		if (state->schemaObj && container_type != JSON_LIST) {
			if (state->tok->type == JSMN_STRING || state->tok->type == JSMN_PRIMITIVE) {
				json_decode_add_schema_type(interp, state, json_decode_get_type(state->tok, state->json));
			}
		}

		/* Dump the array or object value */
		json_decode_dump_value(interp, state, list);
	}

	return list;
}

/**
 * Appends the value at state->tok to 'list' and increments state->tok to just
 * past that token.
 *
 * Also appends to the schema if state->enable_schema is set.
 */
static void
json_decode_dump_value(Jim_Interp *interp, struct json_state *state, Jim_Obj *list)
{
	const jsmntok_t *t = state->tok;

	if (t->type == JSMN_STRING || t->type == JSMN_PRIMITIVE) {
		Jim_Obj	*elem;
		int len = t->end - t->start;
		const char *p = state->json + t->start;
		if (t->type == JSMN_STRING) {
			/* Do we need to process backslash escapes? */
			if (state->need_subst == 0 && memchr(p, '\\', len) != NULL) {
				state->need_subst = 1;
			}
			elem = Jim_NewStringObj(interp, p, len);
		} else if (p[0] == 'n') {	/* null */
			elem = state->nullObj;
		} else if (p[0] == 'I') {
			elem = Jim_NewStringObj(interp, "Inf", -1);
		} else if (p[0] == '-' && p[1] == 'I') {
			elem = Jim_NewStringObj(interp, "-Inf", -1);
		} else {		/* number, true or false */
			elem = Jim_NewStringObj(interp, p, len);
		}

		Jim_ListAppendElement(interp, list, elem);
		state->tok++;
	}
	else {
		Jim_Obj *prevSchemaObj = json_decode_schema_push(interp, state);
		Jim_Obj *newList = json_decode_dump_container(interp, state);
		Jim_ListAppendElement(interp, list, newList);
		json_decode_schema_pop(interp, state, prevSchemaObj);
	}
}

/* Parses the options ?-null string? ?-schema? *state.
 * Any options not present are not set.
 *
 * Returns JIM_OK or JIM_ERR and sets an error result.
 */
static int parse_json_decode_options(Jim_Interp *interp, int argc, Jim_Obj *const argv[], struct json_state *state)
{
	static const char * const options[] = { "-index", "-null", "-schema", NULL };
	enum { OPT_INDEX, OPT_NULL, OPT_SCHEMA, };
	int i;

	for (i = 1; i < argc - 1; i++) {
		int option;
		if (Jim_GetEnum(interp, argv[i], options, &option, NULL, JIM_ERRMSG | JIM_ENUM_ABBREV) != JIM_OK) {
			return JIM_ERR;
		}
		switch (option) {
			case OPT_INDEX:
				state->enable_index = 1;
				break;

			case OPT_NULL:
				i++;
				Jim_IncrRefCount(argv[i]);
				Jim_DecrRefCount(interp, state->nullObj);
				state->nullObj = argv[i];
				break;

			case OPT_SCHEMA:
				state->enable_schema = 1;
				break;
		}
	}

	if (i != argc - 1) {
		Jim_WrongNumArgs(interp, 1, argv,
			"?-index? ?-null nullvalue? ?-schema? json");
		return JIM_ERR;
	}

	return JIM_OK;
}

/**
 * Use jsmn to tokenise the JSON string 'json' of length 'len'
 *
 * Returns an allocated array of tokens or NULL on error (and sets an error result)
 */
static jsmntok_t *
json_decode_tokenize(Jim_Interp *interp, const char *json, size_t len)
{
	jsmntok_t	*t;
	jsmn_parser	 parser;
	int n;

	/* Parse once just to find the number of tokens */
	jsmn_init(&parser);
	n = jsmn_parse(&parser, json, len, NULL, 0);

error:
	switch (n) {
		case JSMN_ERROR_INVAL:
			Jim_SetResultString(interp, "invalid JSON string", -1);
			return NULL;

		case JSMN_ERROR_PART:
			Jim_SetResultString(interp, "truncated JSON string", -1);
			return NULL;

		case 0:
			Jim_SetResultString(interp, "root element must be an object or an array", -1);
			return NULL;

		default:
			break;
	}

	if (n < 0) {
		return NULL;
	}

	t = Jim_Alloc(n * sizeof(*t));

	jsmn_init(&parser);
	n = jsmn_parse(&parser, json, len, t, n);
	if (t->type != JSMN_OBJECT && t->type != JSMN_ARRAY) {
		n = 0;
	}
	if (n <= 0) {
		Jim_Free(t);
		goto error;
	}

	return t;
}

/**
 * json::decode returns the decoded data structure.
 *
 * If -schema is specified, returns a list of {data schema}
 */
static int
json_decode(Jim_Interp *interp, int argc, Jim_Obj *const argv[])
{
	Jim_Obj *list;
	jsmntok_t *tokens;
	int len;
	int ret = JIM_ERR;
	struct json_state state;

	memset(&state, 0, sizeof(state));

	state.nullObj = Jim_NewStringObj(interp, "null", -1);
	Jim_IncrRefCount(state.nullObj);

	if (parse_json_decode_options(interp, argc, argv, &state) != JIM_OK) {
		goto done;
	}

	state.json = Jim_GetString(argv[argc - 1], &len);

	if (!len) {
		Jim_SetResultString(interp, "empty JSON string", -1);
		goto done;
	}
	if ((tokens = json_decode_tokenize(interp, state.json, len)) == NULL) {
		goto done;
	}
	state.tok = tokens;
	json_decode_schema_push(interp, &state);

	list = json_decode_dump_container(interp, &state);
	Jim_Free(tokens);
	ret = JIM_OK;

	/* Make sure the refcount doesn't go to 0 during Jim_SubstObj() */
	Jim_IncrRefCount(list);

	if (state.need_subst) {
		/* Subsitute backslashes in the returned dictionary.
		 * Need to be careful of refcounts.
		 * Note that Jim_SubstObj() supports a few more escapes than
		 * JSON requires, but should give the same result for all legal escapes.
		 */
		Jim_Obj *newList;
		Jim_SubstObj(interp, list, &newList, JIM_SUBST_FLAG | JIM_SUBST_NOCMD | JIM_SUBST_NOVAR);
		Jim_IncrRefCount(newList);
		Jim_DecrRefCount(interp, list);
		list = newList;
	}

	if (state.schemaObj) {
		Jim_Obj *resultObj = Jim_NewListObj(interp, NULL, 0);
		Jim_ListAppendElement(interp, resultObj, list);
		Jim_ListAppendElement(interp, resultObj, state.schemaObj);
		Jim_SetResult(interp, resultObj);
		Jim_DecrRefCount(interp, state.schemaObj);
	}
	else {
		Jim_SetResult(interp, list);
	}
	Jim_DecrRefCount(interp, list);

done:
	Jim_DecrRefCount(interp, state.nullObj);

	return ret;
}

int
Jim_jsonInit(Jim_Interp *interp)
{
	Jim_PackageProvideCheck(interp, "json");
	Jim_CreateCommand(interp, "json::decode", json_decode, NULL, NULL);
	/* Load the Tcl implementation of the json encoder if possible */
	Jim_PackageRequire(interp, "jsonencode", 0);
	return JIM_OK;
}