Repository navigation
Expand file tree
/
Copy pathtextutils_internal.h
More file actions
426 lines (353 loc) · 15.3 KB
/
Copy pathtextutils_internal.h
File metadata and controls
426 lines (353 loc) · 15.3 KB
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
/* textutils_internal.h -- non-bioinformatics utility routines for text etc.
Copyright (C) 2016,2018-2020, 2024 Genome Research Ltd.
Author: John Marshall <jm18@sanger.ac.uk>
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE. */
#ifndef HTSLIB_TEXTUTILS_INTERNAL_H
#define HTSLIB_TEXTUTILS_INTERNAL_H
/* N.B. These interfaces may be used by plug-ins */
#include <ctype.h>
#include <stdlib.h>
#include "htslib/kstring.h"
#ifdef __cplusplus
extern "C" {
#endif
/// Decode percent-encoded (URL-encoded) text
/** On input, _dest_ should be a buffer at least the same size as _s_,
and may be equal to _s_ to decode in place. On output, _dest_ will be
NUL-terminated and the number of characters written (not including the
NUL) is stored in _destlen_.
*/
int hts_decode_percent(char *dest, size_t *destlen, const char *s);
/// Return decoded data length given length of base64-encoded text
/** This gives an upper bound, as it overestimates by a byte or two when
the encoded text ends with (possibly omitted) `=` padding characters.
*/
size_t hts_base64_decoded_length(size_t len);
/// Decode base64-encoded data
/** On input, _dest_ should be a sufficient buffer (see `hts_base64_length()`),
and may be equal to _s_ to decode in place. On output, the number of
bytes written is stored in _destlen_.
*/
int hts_decode_base64(char *dest, size_t *destlen, const char *s);
/// Token structure returned by JSON lexing functions
/** Structure is defined in hts_internal.h
*/
typedef struct hts_json_token hts_json_token;
/// Allocate an empty JSON token structure, for use with hts_json_* functions
/** @return An empty token on success; NULL on failure
*/
HTSLIB_EXPORT
hts_json_token *hts_json_alloc_token(void);
/// Free a JSON token
HTSLIB_EXPORT
void hts_json_free_token(hts_json_token *token);
/// Accessor function to get JSON token type
/** @param token Pointer to JSON token
@return Character indicating the token type
Token types correspond to scalar JSON values and selected punctuation
as follows:
- `s` string
- `n` number
- `b` boolean literal
- `.` null literal
- `{`, `}`, `[`, `]` object and array delimiters
- `?` lexing error
- `!` other errors (e.g. out of memory)
- `\0` terminator at end of input
*/
HTSLIB_EXPORT
char hts_json_token_type(hts_json_token *token);
/// Accessor function to get JSON token in string form
/** @param token Pointer to JSON token
@return String representation of the JSON token; NULL if unset
If the token was parsed from a string using hts_json_snext(), the return value
will point into the string passed as the first parameter to hts_json_snext().
If the token was parsed from a file using hts_json_fnext(), the return value
will point at the kstring_t buffer passed as the third parameter to
hts_json_fnext(). In that case, the value will only be valid until the
next call to hts_json_fnext().
*/
HTSLIB_EXPORT
char *hts_json_token_str(hts_json_token *token);
/// Read one JSON token from a string
/** @param str The input C string
@param state The input string state
@param token On return, filled in with the token read
@return The type of the token read
On return, `token->str` points into the supplied input string, which
is modified by having token-terminating characters overwritten as NULs.
The `state` argument records the current position within `str` after each
`hts_json_snext()` call, and should be set to 0 before the first call.
*/
HTSLIB_EXPORT
char hts_json_snext(char *str, size_t *state, hts_json_token *token);
/// Read and discard a complete JSON value from a string
/** @param str The input C string
@param state The input string state, as per `hts_json_snext()`
@param type If the first token of the value to be discarded has already
been read, provide its type; otherwise `'\0'`
@return One of `v` (success), `\0` (end of string), and `?` (lexing error)
Skips a complete JSON value, which may be a single token or an entire object
or array.
*/
HTSLIB_EXPORT
char hts_json_sskip_value(char *str, size_t *state, char type);
struct hFILE;
/// Read one JSON token from a file
/** @param fp The file stream
@param token On return, filled in with the token read
@param kstr Buffer used to store the token string returned
@return The type of the token read
The `kstr` buffer is used to store the string value of the token read,
so `token->str` is only valid until the next time `hts_json_fnext()` is
called with the same `kstr` argument.
*/
HTSLIB_EXPORT
char hts_json_fnext(struct hFILE *fp, hts_json_token *token, kstring_t *kstr);
/// Read and discard a complete JSON value from a file
/** @param fp The file stream
@param type If the first token of the value to be discarded has already
been read, provide its type; otherwise `'\0'`
@return One of `v` (success), `\0` (EOF), and `?` (lexing error)
Skips a complete JSON value, which may be a single token or an entire object
or array.
*/
HTSLIB_EXPORT
char hts_json_fskip_value(struct hFILE *fp, char type);
// The <ctype.h> functions operate on ints such as are returned by fgetc(),
// i.e., characters represented as unsigned-char-valued ints, or EOF.
// To operate on plain chars (and to avoid warnings on some platforms),
// technically one must cast to unsigned char everywhere (see CERT STR37-C)
// or less painfully use these *_c() functions that operate on plain chars
// (but not EOF, which must be considered separately where it is applicable).
// TODO We may eventually wish to implement these functions directly without
// using their <ctype.h> equivalents, and thus make them immune to locales.
static inline int isalnum_c(char c) { return isalnum((unsigned char) c); }
static inline int isalpha_c(char c) { return isalpha((unsigned char) c); }
static inline int isdigit_c(char c) { return isdigit((unsigned char) c); }
static inline int isgraph_c(char c) { return isgraph((unsigned char) c); }
static inline int islower_c(char c) { return islower((unsigned char) c); }
static inline int isprint_c(char c) { return isprint((unsigned char) c); }
static inline int ispunct_c(char c) { return ispunct((unsigned char) c); }
static inline int isspace_c(char c) { return isspace((unsigned char) c); }
static inline int isupper_c(char c) { return isupper((unsigned char) c); }
static inline int isxdigit_c(char c) { return isxdigit((unsigned char) c); }
static inline char tolower_c(char c) { return tolower((unsigned char) c); }
static inline char toupper_c(char c) { return toupper((unsigned char) c); }
/// Copy possibly malicious text data to a buffer
/** @param buf Destination buffer
@param buflen Size of the destination buffer (>= 4; >= 6 when quotes used)
@param quote Quote character (or '\0' for no quoting of the output)
@param s String to be copied
@param len Length of the input string, or SIZE_MAX to copy until '\0'
@return The destination buffer, @a buf.
Copies the source text string (escaping any unprintable characters) to the
destination buffer. The destination buffer will always be NUL-terminated;
the text will be truncated (and "..." appended) if necessary to make it fit.
*/
const char *hts_strprint(char *buf, size_t buflen, char quote,
const char *s, size_t len);
// Faster replacements for strtol, for use when parsing lots of numbers.
// Note that these only handle base 10 and do not skip leading whitespace
/// Convert a string to a signed integer, with overflow detection
/** @param[in] in Input string
@param[out] end Returned end pointer
@param[in] bits Bits available for the converted value
@param[out] failed Location of overflow flag
@return String value converted to an int64_t
Converts a signed decimal string to an int64_t. The string should
consist of an optional '+' or '-' sign followed by one or more of
the digits 0 to 9. The output value will be limited to fit in the
given number of bits (including the sign bit). If the value is too big,
the largest possible value will be returned and *failed will be set to 1.
The address of the first character following the converted number will
be stored in *end.
Both end and failed must be non-NULL.
*/
static inline int64_t hts_str2int(const char *in, char **end, int bits,
int *failed) {
uint64_t n = 0, limit = (1ULL << (bits - 1)) - 1;
uint32_t fast = (bits - 1) * 1000 / 3322 + 1; // log(10)/log(2) ~= 3.322
const unsigned char *v = (const unsigned char *) in;
const unsigned int ascii_zero = '0'; // Prevents conversion to signed
unsigned int d;
int neg;
switch(*v) {
case '-':
limit++;
neg=1;
v++;
// See "dup" comment below
while (--fast && *v>='0' && *v<='9')
n = n*10 + *v++ - ascii_zero;
break;
case '+':
v++;
// fall through
default:
neg = 0;
// dup of above. This is somewhat unstable and mainly for code
// size cheats to prevent instruction cache lines spanning 32-byte
// blocks in the sam_parse_B_vals calling code. It's been tested
// on gcc7, gcc13, clang10 and clang16 with -O2 and -O3. While
// not exhaustive, this code duplication gives stable fast results
// while a single copy does not.
// (NB: system was "seq4d", so quite old)
while (--fast && *v>='0' && *v<='9')
n = n*10 + *v++ - ascii_zero;
break;
}
// NB gcc7 is slow with (unsigned)(*v - ascii_zero) < 10,
// while gcc13 prefers it.
if (*v>='0' && !fast) { // rejects ',' and tab
uint64_t limit_d_10 = limit / 10;
uint64_t limit_m_10 = limit - 10 * limit_d_10;
while ((d = *v - ascii_zero) < 10) {
if (n < limit_d_10 || (n == limit_d_10 && d <= limit_m_10)) {
n = n*10 + d;
v++;
} else {
do { v++; } while (*v - ascii_zero < 10);
n = limit;
*failed = 1;
break;
}
}
}
*end = (char *)v;
return neg ? (int64_t)-n : (int64_t)n;
}
/// Convert a string to an unsigned integer, with overflow detection
/** @param[in] in Input string
@param[out] end Returned end pointer
@param[in] bits Bits available for the converted value
@param[out] failed Location of overflow flag
@return String value converted to a uint64_t
Converts an unsigned decimal string to a uint64_t. The string should
consist of an optional '+' sign followed by one or more of the digits 0
to 9. The output value will be limited to fit in the given number of bits.
If the value is too big, the largest possible value will be returned
and *failed will be set to 1.
The address of the first character following the converted number will
be stored in *end.
Both end and failed must be non-NULL.
*/
static inline uint64_t hts_str2uint(const char *in, char **end, int bits,
int *failed) {
uint64_t n = 0, limit = (bits < 64 ? (1ULL << bits) : 0) - 1;
const unsigned char *v = (const unsigned char *) in;
const unsigned int ascii_zero = '0'; // Prevents conversion to signed
uint32_t fast = bits * 1000 / 3322 + 1; // log(10)/log(2) ~= 3.322
unsigned int d;
if (*v == '+')
v++;
while (--fast && *v>='0' && *v<='9')
n = n*10 + *v++ - ascii_zero;
if ((unsigned)(*v - ascii_zero) < 10 && !fast) {
uint64_t limit_d_10 = limit / 10;
uint64_t limit_m_10 = limit - 10 * limit_d_10;
while ((d = *v - ascii_zero) < 10) {
if (n < limit_d_10 || (n == limit_d_10 && d <= limit_m_10)) {
n = n*10 + d;
v++;
} else {
do { v++; } while (*v - ascii_zero < 10);
n = limit;
*failed = 1;
break;
}
}
}
*end = (char *)v;
return n;
}
/// Convert a string to a double, with overflow detection
/** @param[in] in Input string
@param[out] end Returned end pointer
@param[out] failed Location of overflow flag
@return String value converted to a double
Converts a floating point value string to a double. The string should
have the format [+-]?[0-9]*[.]?[0-9]* with at least one and no more than 15
digits. Strings that do not match (inf, nan, values with exponents) will
be passed on to strtod() for processing.
If the value is too big, the largest possible value will be returned;
if it is too small to be represented in a double zero will be returned.
In both cases errno will be set to ERANGE.
If no characters could be converted, *failed will be set to 1.
The address of the first character following the converted number will
be stored in *end.
Both end and failed must be non-NULL.
*/
static inline double hts_str2dbl(const char *in, char **end, int *failed) {
uint64_t n = 0;
int max_len = 15;
const unsigned char *v = (const unsigned char *) in;
const unsigned int ascii_zero = '0'; // Prevents conversion to signed
int neg = 0, point = -1;
double d;
static double D[] = {1,1, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7,
1e8, 1e9, 1e10,1e11,1e12,1e13,1e14,1e15,
1e16,1e17,1e18,1e19,1e20};
while (isspace(*v))
v++;
if (*v == '-') {
neg = 1;
v++;
} else if (*v == '+') {
v++;
}
switch(*v) {
case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
break;
case '0':
if (v[1] != 'x' && v[1] != 'X') break;
// else fall through - hex number
default:
// Non numbers, like NaN, Inf
d = strtod(in, end);
if (*end == in)
*failed = 1;
return d;
}
while (*v == '0') ++v;
const unsigned char *start = v;
while (--max_len && *v>='0' && *v<='9')
n = n*10 + *v++ - ascii_zero;
if (max_len && *v == '.') {
point = v - start;
v++;
while (--max_len && *v>='0' && *v<='9')
n = n*10 + *v++ - ascii_zero;
}
if (point < 0)
point = v - start;
// Outside the scope of this quick and dirty parser.
if (!max_len || *v == 'e' || *v == 'E') {
d = strtod(in, end);
if (*end == in)
*failed = 1;
return d;
}
*end = (char *)v;
d = n / D[v - start - point];
return neg ? -d : d;
}
#ifdef __cplusplus
}
#endif
#endif