aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/builtin_printf.cpp
blob: 9f3ea7bc5cd539c3612f91b352b104026736daca (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
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
// printf - format and print data
// Copyright (C) 1990-2007 Free Software Foundation, Inc.
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 2, or (at your option)
// any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software Foundation,
// Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.  */

// Usage: printf format [argument...]
//
// A front end to the printf function that lets it be used from the shell.
//
// Backslash escapes:
//
// \" = double quote
// \\ = backslash
// \a = alert (bell)
// \b = backspace
// \c = produce no further output
// \e = escape
// \f = form feed
// \n = new line
// \r = carriage return
// \t = horizontal tab
// \v = vertical tab
// \ooo = octal number (ooo is 1 to 3 digits)
// \xhh = hexadecimal number (hhh is 1 to 2 digits)
// \uhhhh = 16-bit Unicode character (hhhh is 4 digits)
// \Uhhhhhhhh = 32-bit Unicode character (hhhhhhhh is 8 digits)
//
// Additional directive:
//
// %b = print an argument string, interpreting backslash escapes,
//   except that octal escapes are of the form \0 or \0ooo.
//
// The `format' argument is re-used as many times as necessary
// to convert all of the given arguments.
//
// David MacKenzie <djm@gnu.ai.mit.edu>

// This file has been imported from source code of printf command in GNU Coreutils version 6.9.
#include "config.h"  // IWYU pragma: keep

#include <errno.h>
#include <limits.h>
#include <locale.h>
#include <stdarg.h>
#include <stdio.h>
#include <string.h>
#include <sys/types.h>
#include <wchar.h>
#include <wctype.h>

#include "builtin.h"
#include "common.h"
#include "io.h"
#include "proc.h"
#include "wutil.h"  // IWYU pragma: keep

class parser_t;

struct builtin_printf_state_t {
    // Out and err streams. Note this is a captured reference!
    io_streams_t &streams;

    // The status of the operation.
    int exit_code;

    // Whether we should stop outputting. This gets set in the case of an error, and also with the
    // \c escape.
    bool early_exit;

    explicit builtin_printf_state_t(io_streams_t &s)
        : streams(s), exit_code(0), early_exit(false) {}

    void verify_numeric(const wchar_t *s, const wchar_t *end, int errcode);

    void print_direc(const wchar_t *start, size_t length, wchar_t conversion, bool have_field_width,
                     int field_width, bool have_precision, int precision, wchar_t const *argument);

    int print_formatted(const wchar_t *format, int argc, wchar_t **argv);

    void fatal_error(const wchar_t *format, ...);

    long print_esc(const wchar_t *escstart, bool octal_0);
    void print_esc_string(const wchar_t *str);
    void print_esc_char(wchar_t c);

    void append_output(wchar_t c);
    void append_output(const wchar_t *c);
    void append_format_output(const wchar_t *fmt, ...);
};

static bool is_octal_digit(wchar_t c) { return c != L'\0' && wcschr(L"01234567", c) != NULL; }

static bool is_hex_digit(wchar_t c) {
    return c != L'\0' && wcschr(L"0123456789ABCDEFabcdef", c) != NULL;
}

static int hex_to_bin(const wchar_t &c) {
    switch (c) {
        case L'0': {
            return 0;
        }
        case L'1': {
            return 1;
        }
        case L'2': {
            return 2;
        }
        case L'3': {
            return 3;
        }
        case L'4': {
            return 4;
        }
        case L'5': {
            return 5;
        }
        case L'6': {
            return 6;
        }
        case L'7': {
            return 7;
        }
        case L'8': {
            return 8;
        }
        case L'9': {
            return 9;
        }
        case L'a':
        case L'A': {
            return 10;
        }
        case L'b':
        case L'B': {
            return 11;
        }
        case L'c':
        case L'C': {
            return 12;
        }
        case L'd':
        case L'D': {
            return 13;
        }
        case L'e':
        case L'E': {
            return 14;
        }
        case L'f':
        case L'F': {
            return 15;
        }
        default: { return -1; }
    }
}

static int octal_to_bin(wchar_t c) {
    switch (c) {
        case L'0': {
            return 0;
        }
        case L'1': {
            return 1;
        }
        case L'2': {
            return 2;
        }
        case L'3': {
            return 3;
        }
        case L'4': {
            return 4;
        }
        case L'5': {
            return 5;
        }
        case L'6': {
            return 6;
        }
        case L'7': {
            return 7;
        }
        default: { return -1; }
    }
}

void builtin_printf_state_t::fatal_error(const wchar_t *fmt, ...) {
    // Don't error twice.
    if (early_exit) return;

    va_list va;
    va_start(va, fmt);
    wcstring errstr = vformat_string(fmt, va);
    va_end(va);
    streams.err.append(errstr);
    if (!string_suffixes_string(L"\n", errstr)) streams.err.push_back(L'\n');

    this->exit_code = STATUS_BUILTIN_ERROR;
    this->early_exit = true;
}

void builtin_printf_state_t::append_output(wchar_t c) {
    // Don't output if we're done.
    if (early_exit) return;

    streams.out.push_back(c);
}

void builtin_printf_state_t::append_output(const wchar_t *c) {
    // Don't output if we're done.
    if (early_exit) return;

    streams.out.append(c);
}

void builtin_printf_state_t::append_format_output(const wchar_t *fmt, ...) {
    // Don't output if we're done.
    if (early_exit) return;

    va_list va;
    va_start(va, fmt);
    wcstring tmp = vformat_string(fmt, va);
    va_end(va);
    streams.out.append(tmp);
}

void builtin_printf_state_t::verify_numeric(const wchar_t *s, const wchar_t *end, int errcode) {
    if (errcode != 0) {
        this->fatal_error(L"%ls: %s", s, strerror(errcode));
    } else if (*end) {
        if (s == end)
            this->fatal_error(_(L"%ls: expected a numeric value"), s);
        else
            this->fatal_error(_(L"%ls: value not completely converted"), s);
    }
}

template <typename T>
static T raw_string_to_scalar_type(const wchar_t *s, wchar_t **end);

// we use wcstoll instead of wcstoimax because FreeBSD 8 has busted wcstoumax and wcstoimax - see
// #626
template <>
intmax_t raw_string_to_scalar_type(const wchar_t *s, wchar_t **end) {
    return wcstoll(s, end, 0);
}

template <>
uintmax_t raw_string_to_scalar_type(const wchar_t *s, wchar_t **end) {
    return wcstoull(s, end, 0);
}

template <>
long double raw_string_to_scalar_type(const wchar_t *s, wchar_t **end) {
    // Forcing the locale to C is questionable but it's what the old C_STRTOD() that I inlined here
    // as part of changing how locale management is done by fish.
    char *old_locale = setlocale(LC_NUMERIC, "C");
    double val = wcstod(s, end);
    setlocale(LC_NUMERIC, old_locale);
    return val;
}

template <typename T>
static T string_to_scalar_type(const wchar_t *s, builtin_printf_state_t *state) {
    T val;
    if (*s == L'\"' || *s == L'\'') {
        wchar_t ch = *++s;
        val = ch;
    } else {
        wchar_t *end = NULL;
        errno = 0;
        val = raw_string_to_scalar_type<T>(s, &end);
        state->verify_numeric(s, end, errno);
    }
    return val;
}

/// Output a single-character \ escape.
void builtin_printf_state_t::print_esc_char(wchar_t c) {
    switch (c) {
        case L'a': {  // alert
            this->append_output(L'\a');
            break;
        }
        case L'b': {  // backspace
            this->append_output(L'\b');
            break;
        }
        case L'c': {  // cancel the rest of the output
            this->early_exit = true;
            break;
        }
        case L'e': {  // escape
            this->append_output(L'\x1B');
            break;
        }
        case L'f': {  // form feed
            this->append_output(L'\f');
            break;
        }
        case L'n': {  // new line
            this->append_output(L'\n');
            break;
        }
        case L'r': {  // carriage return
            this->append_output(L'\r');
            break;
        }
        case L't': {  // horizontal tab
            this->append_output(L'\t');
            break;
        }
        case L'v': {  // vertical tab
            this->append_output(L'\v');
            break;
        }
        default: {
            this->append_output(c);
            break;
        }
    }
}

/// Print a \ escape sequence starting at ESCSTART.
/// Return the number of characters in the escape sequence besides the backslash..
/// If OCTAL_0 is nonzero, octal escapes are of the form \0ooo, where o
/// is an octal digit; otherwise they are of the form \ooo.
long builtin_printf_state_t::print_esc(const wchar_t *escstart, bool octal_0) {
    const wchar_t *p = escstart + 1;
    int esc_value = 0; /* Value of \nnn escape. */
    int esc_length;    /* Length of \nnn escape. */

    if (*p == L'x') {
        // A hexadecimal \xhh escape sequence must have 1 or 2 hex. digits.
        for (esc_length = 0, ++p; esc_length < 2 && is_hex_digit(*p); ++esc_length, ++p)
            esc_value = esc_value * 16 + hex_to_bin(*p);
        if (esc_length == 0) this->fatal_error(_(L"missing hexadecimal number in escape"));
        this->append_output(ENCODE_DIRECT_BASE + esc_value % 256);
    } else if (is_octal_digit(*p)) {
        // Parse \0ooo (if octal_0 && *p == L'0') or \ooo (otherwise). Allow \ooo if octal_0 && *p
        // != L'0'; this is an undocumented extension to POSIX that is compatible with Bash 2.05b.
        // Wrap mod 256, which matches historic behavior.
        for (esc_length = 0, p += octal_0 && *p == L'0'; esc_length < 3 && is_octal_digit(*p);
             ++esc_length, ++p)
            esc_value = esc_value * 8 + octal_to_bin(*p);
        this->append_output(ENCODE_DIRECT_BASE + esc_value % 256);
    } else if (*p && wcschr(L"\"\\abcefnrtv", *p)) {
        print_esc_char(*p++);
    } else if (*p == L'u' || *p == L'U') {
        wchar_t esc_char = *p;
        p++;
        uint32_t uni_value = 0;
        for (size_t esc_length = 0; esc_length < (esc_char == L'u' ? 4 : 8); esc_length++) {
            if (!is_hex_digit(*p)) {
                // Escape sequence must be done. Complain if we didn't get anything.
                if (esc_length == 0) {
                    this->fatal_error(_(L"Missing hexadecimal number in Unicode escape"));
                }
                break;
            }
            uni_value = uni_value * 16 + hex_to_bin(*p);
            p++;
        }

        // PCA GNU printf respects the limitations described in ISO N717, about which universal
        // characters "shall not" be specified. I believe this limitation is for the benefit of
        // compilers; I see no reason to impose it in builtin_printf.
        //
        // If __STDC_ISO_10646__ is defined, then it means wchar_t can and does hold Unicode code
        // points, so just use that. If not defined, use the %lc printf conversion; this probably
        // won't do anything good if your wide character set is not Unicode, but such platforms are
        // exceedingly rare.
        if (uni_value > 0x10FFFF) {
            this->fatal_error(_(L"Unicode character out of range: \\%c%0*x"), esc_char,
                              (esc_char == L'u' ? 4 : 8), uni_value);
        } else {
#if defined(__STDC_ISO_10646__)
            this->append_output(uni_value);
#else
            this->append_format_output(L"%lc", uni_value);
#endif
        }
    } else {
        this->append_output(L'\\');
        if (*p) {
            this->append_output(*p);
            p++;
        }
    }
    return p - escstart - 1;
}

/// Print string STR, evaluating \ escapes.
void builtin_printf_state_t::print_esc_string(const wchar_t *str) {
    for (; *str; str++)
        if (*str == L'\\')
            str += print_esc(str, true);
        else
            this->append_output(*str);
}

/// Evaluate a printf conversion specification.  START is the start of the directive, LENGTH is its
/// length, and CONVERSION specifies the type of conversion.  LENGTH does not include any length
/// modifier or the conversion specifier itself.  FIELD_WIDTH and PRECISION are the field width and
/// precision for '*' values, if HAVE_FIELD_WIDTH and HAVE_PRECISION are true, respectively.
/// ARGUMENT is the argument to be formatted.
void builtin_printf_state_t::print_direc(const wchar_t *start, size_t length, wchar_t conversion,
                                         bool have_field_width, int field_width,
                                         bool have_precision, int precision,
                                         wchar_t const *argument) {
    // Start with everything except the conversion specifier.
    wcstring fmt(start, length);

    // Create a copy of the % directive, with an intmax_t-wide width modifier substituted for any
    // existing integer length modifier.
    switch (conversion) {
        case L'd':
        case L'i':
        case L'u': {
            fmt.append(L"ll");
            break;
        }
        case L'a':
        case L'e':
        case L'f':
        case L'g':
        case L'A':
        case L'E':
        case L'F':
        case L'G': {
            fmt.append(L"L");
            break;
        }
        case L's':
        case L'c': {
            fmt.append(L"l");
            break;
        }
        default: { break; }
    }

    // Append the conversion itself.
    fmt.push_back(conversion);

    switch (conversion) {
        case L'd':
        case L'i': {
            intmax_t arg = string_to_scalar_type<intmax_t>(argument, this);
            if (!have_field_width) {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), arg);
                else
                    this->append_format_output(fmt.c_str(), precision, arg);
            } else {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), field_width, arg);
                else
                    this->append_format_output(fmt.c_str(), field_width, precision, arg);
            }
            break;
        }
        case L'o':
        case L'u':
        case L'x':
        case L'X': {
            uintmax_t arg = string_to_scalar_type<uintmax_t>(argument, this);
            if (!have_field_width) {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), arg);
                else
                    this->append_format_output(fmt.c_str(), precision, arg);
            } else {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), field_width, arg);
                else
                    this->append_format_output(fmt.c_str(), field_width, precision, arg);
            }
            break;
        }
        case L'a':
        case L'A':
        case L'e':
        case L'E':
        case L'f':
        case L'F':
        case L'g':
        case L'G': {
            long double arg = string_to_scalar_type<long double>(argument, this);
            if (!have_field_width) {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), arg);
                else
                    this->append_format_output(fmt.c_str(), precision, arg);
            } else {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), field_width, arg);
                else
                    this->append_format_output(fmt.c_str(), field_width, precision, arg);
            }
            break;
        }
        case L'c': {
            if (!have_field_width)
                this->append_format_output(fmt.c_str(), *argument);
            else
                this->append_format_output(fmt.c_str(), field_width, *argument);
            break;
        }
        case L's': {
            if (!have_field_width) {
                if (!have_precision) {
                    this->append_format_output(fmt.c_str(), argument);
                } else
                    this->append_format_output(fmt.c_str(), precision, argument);
            } else {
                if (!have_precision)
                    this->append_format_output(fmt.c_str(), field_width, argument);
                else
                    this->append_format_output(fmt.c_str(), field_width, precision, argument);
            }
            break;
        }
    }
}

/// For each character in str, set the corresponding boolean in the array to the given flag.
static inline void modify_allowed_format_specifiers(bool ok[UCHAR_MAX + 1], const char *str,
                                                    bool flag) {
    for (const char *c = str; *c != '\0'; c++) {
        unsigned char idx = static_cast<unsigned char>(*c);
        ok[idx] = flag;
    }
}

/// Print the text in FORMAT, using ARGV (with ARGC elements) for arguments to any `%' directives.
/// Return the number of elements of ARGV used.
int builtin_printf_state_t::print_formatted(const wchar_t *format, int argc, wchar_t **argv) {
    int save_argc = argc;        /* Preserve original value.  */
    const wchar_t *f;            /* Pointer into `format'.  */
    const wchar_t *direc_start;  /* Start of % directive.  */
    size_t direc_length;         /* Length of % directive.  */
    bool have_field_width;       /* True if FIELD_WIDTH is valid.  */
    int field_width = 0;         /* Arg to first '*'.  */
    bool have_precision;         /* True if PRECISION is valid.  */
    int precision = 0;           /* Arg to second '*'.  */
    bool ok[UCHAR_MAX + 1] = {}; /* ok['x'] is true if %x is allowed.  */

    for (f = format; *f != L'\0'; ++f) {
        switch (*f) {
            case L'%': {
                direc_start = f++;
                direc_length = 1;
                have_field_width = have_precision = false;
                if (*f == L'%') {
                    this->append_output(L'%');
                    break;
                }
                if (*f == L'b') {
                    // FIXME: Field width and precision are not supported for %b, even though POSIX
                    // requires it.
                    if (argc > 0) {
                        print_esc_string(*argv);
                        ++argv;
                        --argc;
                    }
                    break;
                }

                modify_allowed_format_specifiers(ok, "aAcdeEfFgGiosuxX", true);

                for (;; f++, direc_length++) {
                    switch (*f) {
                        case L'I':
                        case L'\'': {
                            modify_allowed_format_specifiers(ok, "aAceEosxX", false);
                            break;
                        }
                        case '-':
                        case '+':
                        case ' ': {
                            break;
                        }
                        case L'#': {
                            modify_allowed_format_specifiers(ok, "cdisu", false);
                            break;
                        }
                        case '0': {
                            modify_allowed_format_specifiers(ok, "cs", false);
                            break;
                        }
                        default: { goto no_more_flag_characters; }
                    }
                }
            no_more_flag_characters:;

                if (*f == L'*') {
                    ++f;
                    ++direc_length;
                    if (argc > 0) {
                        intmax_t width = string_to_scalar_type<intmax_t>(*argv, this);
                        if (INT_MIN <= width && width <= INT_MAX)
                            field_width = static_cast<int>(width);
                        else
                            this->fatal_error(_(L"invalid field width: %ls"), *argv);
                        ++argv;
                        --argc;
                    } else {
                        field_width = 0;
                    }
                    have_field_width = true;
                } else {
                    while (iswdigit(*f)) {
                        ++f;
                        ++direc_length;
                    }
                }
                if (*f == L'.') {
                    ++f;
                    ++direc_length;
                    modify_allowed_format_specifiers(ok, "c", false);
                    if (*f == L'*') {
                        ++f;
                        ++direc_length;
                        if (argc > 0) {
                            intmax_t prec = string_to_scalar_type<intmax_t>(*argv, this);
                            if (prec < 0) {
                                // A negative precision is taken as if the precision were omitted,
                                // so -1 is safe here even if prec < INT_MIN.
                                precision = -1;
                            } else if (INT_MAX < prec)
                                this->fatal_error(_(L"invalid precision: %ls"), *argv);
                            else {
                                precision = static_cast<int>(prec);
                            }
                            ++argv;
                            --argc;
                        } else {
                            precision = 0;
                        }
                        have_precision = true;
                    } else {
                        while (iswdigit(*f)) {
                            ++f;
                            ++direc_length;
                        }
                    }
                }

                while (*f == L'l' || *f == L'L' || *f == L'h' || *f == L'j' || *f == L't' ||
                       *f == L'z') {
                    ++f;
                }

                wchar_t conversion = *f;
                if (conversion > 0xFF || !ok[conversion]) {
                    this->fatal_error(_(L"%.*ls: invalid conversion specification"),
                                      (int)(f + 1 - direc_start), direc_start);
                    return 0;
                }

                print_direc(direc_start, direc_length, *f, have_field_width, field_width,
                            have_precision, precision, (argc <= 0 ? L"" : (argc--, *argv++)));
                break;
            }
            case L'\\': {
                f += print_esc(f, false);
                break;
            }
            default: { this->append_output(*f); }
        }
    }
    return save_argc - argc;
}

/// The printf builtin.
int builtin_printf(parser_t &parser, io_streams_t &streams, wchar_t **argv) {
    builtin_printf_state_t state(streams);

    wchar_t *format;
    int args_used;
    int argc = builtin_count_args(argv);

    if (argc <= 1) {
        state.fatal_error(_(L"printf: not enough arguments"));
        return STATUS_BUILTIN_ERROR;
    }

    format = argv[1];
    argc -= 2;
    argv += 2;

    do {
        args_used = state.print_formatted(format, argc, argv);
        argc -= args_used;
        argv += args_used;
    } while (args_used > 0 && argc > 0 && !state.early_exit);
    return state.exit_code;
}