yarp 0.9.0 → 0.10.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +15 -1
- data/Makefile +5 -1
- data/config.yml +156 -125
- data/docs/encoding.md +5 -5
- data/docs/serialization.md +2 -2
- data/ext/yarp/api_node.c +142 -98
- data/ext/yarp/extension.c +21 -7
- data/ext/yarp/extension.h +1 -1
- data/include/yarp/ast.h +327 -18
- data/include/yarp/defines.h +2 -1
- data/include/yarp/diagnostic.h +3 -3
- data/include/yarp/enc/yp_encoding.h +10 -10
- data/include/yarp/parser.h +19 -19
- data/include/yarp/regexp.h +1 -1
- data/include/yarp/unescape.h +4 -4
- data/include/yarp/util/yp_buffer.h +3 -0
- data/include/yarp/util/yp_char.h +16 -16
- data/include/yarp/util/yp_constant_pool.h +2 -2
- data/include/yarp/util/yp_newline_list.h +5 -5
- data/include/yarp/util/yp_string.h +4 -4
- data/include/yarp/util/yp_string_list.h +0 -3
- data/include/yarp/util/yp_strpbrk.h +1 -1
- data/include/yarp/version.h +2 -2
- data/include/yarp.h +5 -4
- data/lib/yarp/desugar_visitor.rb +59 -122
- data/lib/yarp/node.rb +230 -240
- data/lib/yarp/serialize.rb +16 -16
- data/lib/yarp.rb +5 -5
- data/src/diagnostic.c +1 -1
- data/src/enc/yp_big5.c +15 -42
- data/src/enc/yp_euc_jp.c +16 -43
- data/src/enc/yp_gbk.c +19 -46
- data/src/enc/yp_shift_jis.c +16 -43
- data/src/enc/yp_tables.c +36 -38
- data/src/enc/yp_unicode.c +20 -25
- data/src/enc/yp_windows_31j.c +16 -43
- data/src/node.c +1271 -899
- data/src/prettyprint.c +87 -48
- data/src/regexp.c +21 -21
- data/src/serialize.c +28 -15
- data/src/unescape.c +151 -121
- data/src/util/yp_buffer.c +7 -2
- data/src/util/yp_char.c +34 -34
- data/src/util/yp_constant_pool.c +4 -4
- data/src/util/yp_memchr.c +1 -1
- data/src/util/yp_newline_list.c +5 -4
- data/src/util/yp_string.c +22 -20
- data/src/util/yp_string_list.c +0 -6
- data/src/util/yp_strncasecmp.c +3 -6
- data/src/util/yp_strpbrk.c +8 -8
- data/src/yarp.c +355 -216
- data/yarp.gemspec +1 -1
- metadata +2 -2
data/src/util/yp_char.c
CHANGED
@@ -13,8 +13,8 @@
|
|
13
13
|
#define YP_NUMBER_BIT_HEXADECIMAL_DIGIT (1 << 6)
|
14
14
|
#define YP_NUMBER_BIT_HEXADECIMAL_NUMBER (1 << 7)
|
15
15
|
|
16
|
-
static const
|
17
|
-
//0 1 2 3 4 5 6 7 8 9 A B C D E F
|
16
|
+
static const uint8_t yp_byte_table[256] = {
|
17
|
+
// 0 1 2 3 4 5 6 7 8 9 A B C D E F
|
18
18
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 1, 3, 3, 3, 0, 0, // 0x
|
19
19
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 1x
|
20
20
|
3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 2x
|
@@ -33,7 +33,7 @@ static const unsigned char yp_char_table[256] = {
|
|
33
33
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // Fx
|
34
34
|
};
|
35
35
|
|
36
|
-
static const
|
36
|
+
static const uint8_t yp_number_table[256] = {
|
37
37
|
// 0 1 2 3 4 5 6 7 8 9 A B C D E F
|
38
38
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 0x
|
39
39
|
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, // 1x
|
@@ -54,20 +54,20 @@ static const unsigned char yp_number_table[256] = {
|
|
54
54
|
};
|
55
55
|
|
56
56
|
static inline size_t
|
57
|
-
yp_strspn_char_kind(const
|
57
|
+
yp_strspn_char_kind(const uint8_t *string, ptrdiff_t length, uint8_t kind) {
|
58
58
|
if (length <= 0) return 0;
|
59
59
|
|
60
60
|
size_t size = 0;
|
61
61
|
size_t maximum = (size_t) length;
|
62
62
|
|
63
|
-
while (size < maximum && (
|
63
|
+
while (size < maximum && (yp_byte_table[string[size]] & kind)) size++;
|
64
64
|
return size;
|
65
65
|
}
|
66
66
|
|
67
67
|
// Returns the number of characters at the start of the string that are
|
68
68
|
// whitespace. Disallows searching past the given maximum number of characters.
|
69
69
|
size_t
|
70
|
-
yp_strspn_whitespace(const
|
70
|
+
yp_strspn_whitespace(const uint8_t *string, ptrdiff_t length) {
|
71
71
|
return yp_strspn_char_kind(string, length, YP_CHAR_BIT_WHITESPACE);
|
72
72
|
}
|
73
73
|
|
@@ -75,13 +75,13 @@ yp_strspn_whitespace(const char *string, ptrdiff_t length) {
|
|
75
75
|
// whitespace while also tracking the location of each newline. Disallows
|
76
76
|
// searching past the given maximum number of characters.
|
77
77
|
size_t
|
78
|
-
yp_strspn_whitespace_newlines(const
|
78
|
+
yp_strspn_whitespace_newlines(const uint8_t *string, ptrdiff_t length, yp_newline_list_t *newline_list, bool stop_at_newline) {
|
79
79
|
if (length <= 0) return 0;
|
80
80
|
|
81
81
|
size_t size = 0;
|
82
82
|
size_t maximum = (size_t) length;
|
83
83
|
|
84
|
-
while (size < maximum && (
|
84
|
+
while (size < maximum && (yp_byte_table[string[size]] & YP_CHAR_BIT_WHITESPACE)) {
|
85
85
|
if (string[size] == '\n') {
|
86
86
|
if (stop_at_newline) {
|
87
87
|
return size + 1;
|
@@ -100,42 +100,42 @@ yp_strspn_whitespace_newlines(const char *string, ptrdiff_t length, yp_newline_l
|
|
100
100
|
// Returns the number of characters at the start of the string that are inline
|
101
101
|
// whitespace. Disallows searching past the given maximum number of characters.
|
102
102
|
size_t
|
103
|
-
yp_strspn_inline_whitespace(const
|
103
|
+
yp_strspn_inline_whitespace(const uint8_t *string, ptrdiff_t length) {
|
104
104
|
return yp_strspn_char_kind(string, length, YP_CHAR_BIT_INLINE_WHITESPACE);
|
105
105
|
}
|
106
106
|
|
107
107
|
// Returns the number of characters at the start of the string that are regexp
|
108
108
|
// options. Disallows searching past the given maximum number of characters.
|
109
109
|
size_t
|
110
|
-
yp_strspn_regexp_option(const
|
110
|
+
yp_strspn_regexp_option(const uint8_t *string, ptrdiff_t length) {
|
111
111
|
return yp_strspn_char_kind(string, length, YP_CHAR_BIT_REGEXP_OPTION);
|
112
112
|
}
|
113
113
|
|
114
114
|
static inline bool
|
115
|
-
yp_char_is_char_kind(const
|
116
|
-
return (
|
115
|
+
yp_char_is_char_kind(const uint8_t b, uint8_t kind) {
|
116
|
+
return (yp_byte_table[b] & kind) != 0;
|
117
117
|
}
|
118
118
|
|
119
119
|
// Returns true if the given character is a whitespace character.
|
120
120
|
bool
|
121
|
-
yp_char_is_whitespace(const
|
122
|
-
return yp_char_is_char_kind(
|
121
|
+
yp_char_is_whitespace(const uint8_t b) {
|
122
|
+
return yp_char_is_char_kind(b, YP_CHAR_BIT_WHITESPACE);
|
123
123
|
}
|
124
124
|
|
125
125
|
// Returns true if the given character is an inline whitespace character.
|
126
126
|
bool
|
127
|
-
yp_char_is_inline_whitespace(const
|
128
|
-
return yp_char_is_char_kind(
|
127
|
+
yp_char_is_inline_whitespace(const uint8_t b) {
|
128
|
+
return yp_char_is_char_kind(b, YP_CHAR_BIT_INLINE_WHITESPACE);
|
129
129
|
}
|
130
130
|
|
131
131
|
static inline size_t
|
132
|
-
yp_strspn_number_kind(const
|
132
|
+
yp_strspn_number_kind(const uint8_t *string, ptrdiff_t length, uint8_t kind) {
|
133
133
|
if (length <= 0) return 0;
|
134
134
|
|
135
135
|
size_t size = 0;
|
136
136
|
size_t maximum = (size_t) length;
|
137
137
|
|
138
|
-
while (size < maximum && (yp_number_table[
|
138
|
+
while (size < maximum && (yp_number_table[string[size]] & kind)) size++;
|
139
139
|
return size;
|
140
140
|
}
|
141
141
|
|
@@ -143,7 +143,7 @@ yp_strspn_number_kind(const char *string, ptrdiff_t length, unsigned char kind)
|
|
143
143
|
// digits or underscores. Disallows searching past the given maximum number of
|
144
144
|
// characters.
|
145
145
|
size_t
|
146
|
-
yp_strspn_binary_number(const
|
146
|
+
yp_strspn_binary_number(const uint8_t *string, ptrdiff_t length) {
|
147
147
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_BINARY_NUMBER);
|
148
148
|
}
|
149
149
|
|
@@ -151,14 +151,14 @@ yp_strspn_binary_number(const char *string, ptrdiff_t length) {
|
|
151
151
|
// digits or underscores. Disallows searching past the given maximum number of
|
152
152
|
// characters.
|
153
153
|
size_t
|
154
|
-
yp_strspn_octal_number(const
|
154
|
+
yp_strspn_octal_number(const uint8_t *string, ptrdiff_t length) {
|
155
155
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_OCTAL_NUMBER);
|
156
156
|
}
|
157
157
|
|
158
158
|
// Returns the number of characters at the start of the string that are decimal
|
159
159
|
// digits. Disallows searching past the given maximum number of characters.
|
160
160
|
size_t
|
161
|
-
yp_strspn_decimal_digit(const
|
161
|
+
yp_strspn_decimal_digit(const uint8_t *string, ptrdiff_t length) {
|
162
162
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_DECIMAL_DIGIT);
|
163
163
|
}
|
164
164
|
|
@@ -166,7 +166,7 @@ yp_strspn_decimal_digit(const char *string, ptrdiff_t length) {
|
|
166
166
|
// digits or underscores. Disallows searching past the given maximum number of
|
167
167
|
// characters.
|
168
168
|
size_t
|
169
|
-
yp_strspn_decimal_number(const
|
169
|
+
yp_strspn_decimal_number(const uint8_t *string, ptrdiff_t length) {
|
170
170
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_DECIMAL_NUMBER);
|
171
171
|
}
|
172
172
|
|
@@ -174,7 +174,7 @@ yp_strspn_decimal_number(const char *string, ptrdiff_t length) {
|
|
174
174
|
// hexadecimal digits. Disallows searching past the given maximum number of
|
175
175
|
// characters.
|
176
176
|
size_t
|
177
|
-
yp_strspn_hexadecimal_digit(const
|
177
|
+
yp_strspn_hexadecimal_digit(const uint8_t *string, ptrdiff_t length) {
|
178
178
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_HEXADECIMAL_DIGIT);
|
179
179
|
}
|
180
180
|
|
@@ -182,37 +182,37 @@ yp_strspn_hexadecimal_digit(const char *string, ptrdiff_t length) {
|
|
182
182
|
// hexadecimal digits or underscores. Disallows searching past the given maximum
|
183
183
|
// number of characters.
|
184
184
|
size_t
|
185
|
-
yp_strspn_hexadecimal_number(const
|
185
|
+
yp_strspn_hexadecimal_number(const uint8_t *string, ptrdiff_t length) {
|
186
186
|
return yp_strspn_number_kind(string, length, YP_NUMBER_BIT_HEXADECIMAL_NUMBER);
|
187
187
|
}
|
188
188
|
|
189
189
|
static inline bool
|
190
|
-
yp_char_is_number_kind(const
|
191
|
-
return (yp_number_table[
|
190
|
+
yp_char_is_number_kind(const uint8_t b, uint8_t kind) {
|
191
|
+
return (yp_number_table[b] & kind) != 0;
|
192
192
|
}
|
193
193
|
|
194
194
|
// Returns true if the given character is a binary digit.
|
195
195
|
bool
|
196
|
-
yp_char_is_binary_digit(const
|
197
|
-
return yp_char_is_number_kind(
|
196
|
+
yp_char_is_binary_digit(const uint8_t b) {
|
197
|
+
return yp_char_is_number_kind(b, YP_NUMBER_BIT_BINARY_DIGIT);
|
198
198
|
}
|
199
199
|
|
200
200
|
// Returns true if the given character is an octal digit.
|
201
201
|
bool
|
202
|
-
yp_char_is_octal_digit(const
|
203
|
-
return yp_char_is_number_kind(
|
202
|
+
yp_char_is_octal_digit(const uint8_t b) {
|
203
|
+
return yp_char_is_number_kind(b, YP_NUMBER_BIT_OCTAL_DIGIT);
|
204
204
|
}
|
205
205
|
|
206
206
|
// Returns true if the given character is a decimal digit.
|
207
207
|
bool
|
208
|
-
yp_char_is_decimal_digit(const
|
209
|
-
return yp_char_is_number_kind(
|
208
|
+
yp_char_is_decimal_digit(const uint8_t b) {
|
209
|
+
return yp_char_is_number_kind(b, YP_NUMBER_BIT_DECIMAL_DIGIT);
|
210
210
|
}
|
211
211
|
|
212
212
|
// Returns true if the given character is a hexadecimal digit.
|
213
213
|
bool
|
214
|
-
yp_char_is_hexadecimal_digit(const
|
215
|
-
return yp_char_is_number_kind(
|
214
|
+
yp_char_is_hexadecimal_digit(const uint8_t b) {
|
215
|
+
return yp_char_is_number_kind(b, YP_NUMBER_BIT_HEXADECIMAL_DIGIT);
|
216
216
|
}
|
217
217
|
|
218
218
|
#undef YP_CHAR_BIT_WHITESPACE
|
data/src/util/yp_constant_pool.c
CHANGED
@@ -48,12 +48,12 @@ yp_constant_id_list_free(yp_constant_id_list_t *list) {
|
|
48
48
|
// A relatively simple hash function (djb2) that is used to hash strings. We are
|
49
49
|
// optimizing here for simplicity and speed.
|
50
50
|
static inline size_t
|
51
|
-
yp_constant_pool_hash(const
|
51
|
+
yp_constant_pool_hash(const uint8_t *start, size_t length) {
|
52
52
|
// This is a prime number used as the initial value for the hash function.
|
53
53
|
size_t value = 5381;
|
54
54
|
|
55
55
|
for (size_t index = 0; index < length; index++) {
|
56
|
-
value = ((value << 5) + value) +
|
56
|
+
value = ((value << 5) + value) + start[index];
|
57
57
|
}
|
58
58
|
|
59
59
|
return value;
|
@@ -109,7 +109,7 @@ yp_constant_pool_init(yp_constant_pool_t *pool, size_t capacity) {
|
|
109
109
|
// Insert a constant into a constant pool. Returns the id of the constant, or 0
|
110
110
|
// if any potential calls to resize fail.
|
111
111
|
yp_constant_id_t
|
112
|
-
yp_constant_pool_insert(yp_constant_pool_t *pool, const
|
112
|
+
yp_constant_pool_insert(yp_constant_pool_t *pool, const uint8_t *start, size_t length) {
|
113
113
|
if (pool->size >= (pool->capacity / 4 * 3)) {
|
114
114
|
if (!yp_constant_pool_resize(pool)) return 0;
|
115
115
|
}
|
@@ -122,7 +122,7 @@ yp_constant_pool_insert(yp_constant_pool_t *pool, const char *start, size_t leng
|
|
122
122
|
// If there is a collision, then we need to check if the content is the
|
123
123
|
// same as the content we are trying to insert. If it is, then we can
|
124
124
|
// return the id of the existing constant.
|
125
|
-
if ((constant->length == length) &&
|
125
|
+
if ((constant->length == length) && memcmp(constant->start, start, length) == 0) {
|
126
126
|
return pool->constants[index].id;
|
127
127
|
}
|
128
128
|
|
data/src/util/yp_memchr.c
CHANGED
@@ -8,7 +8,7 @@
|
|
8
8
|
void *
|
9
9
|
yp_memchr(const void *memory, int character, size_t number, bool encoding_changed, yp_encoding_t *encoding) {
|
10
10
|
if (encoding_changed && encoding->multibyte && character >= YP_MEMCHR_TRAILING_BYTE_MINIMUM) {
|
11
|
-
const
|
11
|
+
const uint8_t *source = (const uint8_t *) memory;
|
12
12
|
size_t index = 0;
|
13
13
|
|
14
14
|
while (index < number) {
|
data/src/util/yp_newline_list.c
CHANGED
@@ -3,7 +3,7 @@
|
|
3
3
|
// Initialize a new newline list with the given capacity. Returns true if the
|
4
4
|
// allocation of the offsets succeeds, otherwise returns false.
|
5
5
|
bool
|
6
|
-
yp_newline_list_init(yp_newline_list_t *list, const
|
6
|
+
yp_newline_list_init(yp_newline_list_t *list, const uint8_t *start, size_t capacity) {
|
7
7
|
list->offsets = (size_t *) calloc(capacity, sizeof(size_t));
|
8
8
|
if (list->offsets == NULL) return false;
|
9
9
|
|
@@ -23,7 +23,7 @@ yp_newline_list_init(yp_newline_list_t *list, const char *start, size_t capacity
|
|
23
23
|
// Append a new offset to the newline list. Returns true if the reallocation of
|
24
24
|
// the offsets succeeds (if one was necessary), otherwise returns false.
|
25
25
|
bool
|
26
|
-
yp_newline_list_append(yp_newline_list_t *list, const
|
26
|
+
yp_newline_list_append(yp_newline_list_t *list, const uint8_t *cursor) {
|
27
27
|
if (list->size == list->capacity) {
|
28
28
|
list->capacity = (list->capacity * 3) / 2;
|
29
29
|
list->offsets = (size_t *) realloc(list->offsets, list->capacity * sizeof(size_t));
|
@@ -33,6 +33,7 @@ yp_newline_list_append(yp_newline_list_t *list, const char *cursor) {
|
|
33
33
|
assert(*cursor == '\n');
|
34
34
|
assert(cursor >= list->start);
|
35
35
|
size_t newline_offset = (size_t) (cursor - list->start + 1);
|
36
|
+
|
36
37
|
assert(list->size == 0 || newline_offset > list->offsets[list->size - 1]);
|
37
38
|
list->offsets[list->size++] = newline_offset;
|
38
39
|
|
@@ -41,7 +42,7 @@ yp_newline_list_append(yp_newline_list_t *list, const char *cursor) {
|
|
41
42
|
|
42
43
|
// Conditionally append a new offset to the newline list, if the value passed in is a newline.
|
43
44
|
bool
|
44
|
-
yp_newline_list_check_append(yp_newline_list_t *list, const
|
45
|
+
yp_newline_list_check_append(yp_newline_list_t *list, const uint8_t *cursor) {
|
45
46
|
if (*cursor != '\n') {
|
46
47
|
return true;
|
47
48
|
}
|
@@ -105,7 +106,7 @@ yp_newline_list_line_column_scan(yp_newline_list_t *list, size_t offset) {
|
|
105
106
|
// list, the line and column of the closest offset less than the given offset
|
106
107
|
// are returned.
|
107
108
|
yp_line_column_t
|
108
|
-
yp_newline_list_line_column(yp_newline_list_t *list, const
|
109
|
+
yp_newline_list_line_column(yp_newline_list_t *list, const uint8_t *cursor) {
|
109
110
|
assert(cursor >= list->start);
|
110
111
|
size_t offset = (size_t) (cursor - list->start);
|
111
112
|
yp_line_column_t result;
|
data/src/util/yp_string.c
CHANGED
@@ -12,18 +12,19 @@
|
|
12
12
|
|
13
13
|
// Initialize a shared string that is based on initial input.
|
14
14
|
void
|
15
|
-
yp_string_shared_init(yp_string_t *string, const
|
15
|
+
yp_string_shared_init(yp_string_t *string, const uint8_t *start, const uint8_t *end) {
|
16
16
|
assert(start <= end);
|
17
|
+
|
17
18
|
*string = (yp_string_t) {
|
18
19
|
.type = YP_STRING_SHARED,
|
19
|
-
.source =
|
20
|
+
.source = start,
|
20
21
|
.length = (size_t) (end - start)
|
21
22
|
};
|
22
23
|
}
|
23
24
|
|
24
25
|
// Initialize an owned string that is responsible for freeing allocated memory.
|
25
26
|
void
|
26
|
-
yp_string_owned_init(yp_string_t *string,
|
27
|
+
yp_string_owned_init(yp_string_t *string, uint8_t *source, size_t length) {
|
27
28
|
*string = (yp_string_t) {
|
28
29
|
.type = YP_STRING_OWNED,
|
29
30
|
.source = source,
|
@@ -36,13 +37,13 @@ void
|
|
36
37
|
yp_string_constant_init(yp_string_t *string, const char *source, size_t length) {
|
37
38
|
*string = (yp_string_t) {
|
38
39
|
.type = YP_STRING_CONSTANT,
|
39
|
-
.source = (
|
40
|
+
.source = (const uint8_t *) source,
|
40
41
|
.length = length
|
41
42
|
};
|
42
43
|
}
|
43
44
|
|
44
45
|
static void
|
45
|
-
yp_string_mapped_init_internal(yp_string_t *string,
|
46
|
+
yp_string_mapped_init_internal(yp_string_t *string, uint8_t *source, size_t length) {
|
46
47
|
*string = (yp_string_t) {
|
47
48
|
.type = YP_STRING_MAPPED,
|
48
49
|
.source = source,
|
@@ -67,13 +68,13 @@ yp_string_ensure_owned(yp_string_t *string) {
|
|
67
68
|
if (string->type == YP_STRING_OWNED) return;
|
68
69
|
|
69
70
|
size_t length = yp_string_length(string);
|
70
|
-
const
|
71
|
+
const uint8_t *source = yp_string_source(string);
|
71
72
|
|
72
|
-
|
73
|
+
uint8_t *memory = malloc(length);
|
73
74
|
if (!memory) return;
|
74
75
|
|
75
76
|
yp_string_owned_init(string, memory, length);
|
76
|
-
memcpy(string->source, source, length);
|
77
|
+
memcpy((void *) string->source, source, length);
|
77
78
|
}
|
78
79
|
|
79
80
|
// Returns the length associated with the string.
|
@@ -83,7 +84,7 @@ yp_string_length(const yp_string_t *string) {
|
|
83
84
|
}
|
84
85
|
|
85
86
|
// Returns the start pointer associated with the string.
|
86
|
-
YP_EXPORTED_FUNCTION const
|
87
|
+
YP_EXPORTED_FUNCTION const uint8_t *
|
87
88
|
yp_string_source(const yp_string_t *string) {
|
88
89
|
return string->source;
|
89
90
|
}
|
@@ -91,15 +92,16 @@ yp_string_source(const yp_string_t *string) {
|
|
91
92
|
// Free the associated memory of the given string.
|
92
93
|
YP_EXPORTED_FUNCTION void
|
93
94
|
yp_string_free(yp_string_t *string) {
|
95
|
+
void *memory = (void *) string->source;
|
96
|
+
|
94
97
|
if (string->type == YP_STRING_OWNED) {
|
95
|
-
free(
|
98
|
+
free(memory);
|
96
99
|
} else if (string->type == YP_STRING_MAPPED && string->length) {
|
97
|
-
|
98
|
-
#if defined(_WIN32)
|
100
|
+
#if defined(_WIN32)
|
99
101
|
UnmapViewOfFile(memory);
|
100
|
-
|
102
|
+
#else
|
101
103
|
munmap(memory, string->length);
|
102
|
-
|
104
|
+
#endif
|
103
105
|
}
|
104
106
|
}
|
105
107
|
|
@@ -126,8 +128,8 @@ yp_string_mapped_init(yp_string_t *string, const char *filepath) {
|
|
126
128
|
// the source to a constant empty string and return.
|
127
129
|
if (file_size == 0) {
|
128
130
|
CloseHandle(file);
|
129
|
-
|
130
|
-
yp_string_mapped_init_internal(string,
|
131
|
+
uint8_t empty[] = "";
|
132
|
+
yp_string_mapped_init_internal(string, empty, 0);
|
131
133
|
return true;
|
132
134
|
}
|
133
135
|
|
@@ -140,7 +142,7 @@ yp_string_mapped_init(yp_string_t *string, const char *filepath) {
|
|
140
142
|
}
|
141
143
|
|
142
144
|
// Map the file into memory.
|
143
|
-
|
145
|
+
uint8_t *source = (uint8_t *) MapViewOfFile(mapping, FILE_MAP_READ, 0, 0, 0);
|
144
146
|
CloseHandle(mapping);
|
145
147
|
CloseHandle(file);
|
146
148
|
|
@@ -169,12 +171,12 @@ yp_string_mapped_init(yp_string_t *string, const char *filepath) {
|
|
169
171
|
|
170
172
|
// mmap the file descriptor to virtually get the contents
|
171
173
|
size_t size = (size_t) sb.st_size;
|
172
|
-
|
174
|
+
uint8_t *source = NULL;
|
173
175
|
|
174
176
|
if (size == 0) {
|
175
177
|
close(fd);
|
176
|
-
|
177
|
-
yp_string_mapped_init_internal(string,
|
178
|
+
uint8_t empty[] = "";
|
179
|
+
yp_string_mapped_init_internal(string, empty, 0);
|
178
180
|
return true;
|
179
181
|
}
|
180
182
|
|
data/src/util/yp_string_list.c
CHANGED
@@ -1,11 +1,5 @@
|
|
1
1
|
#include "yarp/util/yp_string_list.h"
|
2
2
|
|
3
|
-
// Allocate a new yp_string_list_t.
|
4
|
-
yp_string_list_t *
|
5
|
-
yp_string_list_alloc(void) {
|
6
|
-
return (yp_string_list_t *) malloc(sizeof(yp_string_list_t));
|
7
|
-
}
|
8
|
-
|
9
3
|
// Initialize a yp_string_list_t with its default values.
|
10
4
|
void
|
11
5
|
yp_string_list_init(yp_string_list_t *string_list) {
|
data/src/util/yp_strncasecmp.c
CHANGED
@@ -1,18 +1,15 @@
|
|
1
1
|
#include <ctype.h>
|
2
2
|
#include <stddef.h>
|
3
|
+
#include <stdint.h>
|
3
4
|
|
4
5
|
int
|
5
|
-
yp_strncasecmp(const
|
6
|
+
yp_strncasecmp(const uint8_t *string1, const uint8_t *string2, size_t length) {
|
6
7
|
size_t offset = 0;
|
7
8
|
int difference = 0;
|
8
9
|
|
9
10
|
while (offset < length && string1[offset] != '\0') {
|
10
11
|
if (string2[offset] == '\0') return string1[offset];
|
11
|
-
|
12
|
-
unsigned char left = (unsigned char) string1[offset];
|
13
|
-
unsigned char right = (unsigned char) string2[offset];
|
14
|
-
|
15
|
-
if ((difference = tolower(left) - tolower(right)) != 0) return difference;
|
12
|
+
if ((difference = tolower(string1[offset]) - tolower(string2[offset])) != 0) return difference;
|
16
13
|
offset++;
|
17
14
|
}
|
18
15
|
|
data/src/util/yp_strpbrk.c
CHANGED
@@ -1,12 +1,12 @@
|
|
1
1
|
#include "yarp/util/yp_strpbrk.h"
|
2
2
|
|
3
3
|
// This is the slow path that does care about the encoding.
|
4
|
-
static inline const
|
5
|
-
yp_strpbrk_multi_byte(yp_parser_t *parser, const
|
4
|
+
static inline const uint8_t *
|
5
|
+
yp_strpbrk_multi_byte(yp_parser_t *parser, const uint8_t *source, const uint8_t *charset, size_t maximum) {
|
6
6
|
size_t index = 0;
|
7
7
|
|
8
8
|
while (index < maximum) {
|
9
|
-
if (strchr(charset, source[index]) != NULL) {
|
9
|
+
if (strchr((const char *) charset, source[index]) != NULL) {
|
10
10
|
return source + index;
|
11
11
|
}
|
12
12
|
|
@@ -22,12 +22,12 @@ yp_strpbrk_multi_byte(yp_parser_t *parser, const char *source, const char *chars
|
|
22
22
|
}
|
23
23
|
|
24
24
|
// This is the fast path that does not care about the encoding.
|
25
|
-
static inline const
|
26
|
-
yp_strpbrk_single_byte(const
|
25
|
+
static inline const uint8_t *
|
26
|
+
yp_strpbrk_single_byte(const uint8_t *source, const uint8_t *charset, size_t maximum) {
|
27
27
|
size_t index = 0;
|
28
28
|
|
29
29
|
while (index < maximum) {
|
30
|
-
if (strchr(charset, source[index]) != NULL) {
|
30
|
+
if (strchr((const char *) charset, source[index]) != NULL) {
|
31
31
|
return source + index;
|
32
32
|
}
|
33
33
|
|
@@ -54,8 +54,8 @@ yp_strpbrk_single_byte(const char *source, const char *charset, size_t maximum)
|
|
54
54
|
// characters that are trailing bytes of multi-byte characters. For example, in
|
55
55
|
// Shift-JIS, the backslash character can be a trailing byte. In that case we
|
56
56
|
// need to take a slower path and iterate one multi-byte character at a time.
|
57
|
-
const
|
58
|
-
yp_strpbrk(yp_parser_t *parser, const
|
57
|
+
const uint8_t *
|
58
|
+
yp_strpbrk(yp_parser_t *parser, const uint8_t *source, const uint8_t *charset, ptrdiff_t length) {
|
59
59
|
if (length <= 0) {
|
60
60
|
return NULL;
|
61
61
|
} else if (parser->encoding_changed && parser->encoding.multibyte) {
|