doprnt.c 13.4 KB
Newer Older
Joseph Arceneaux's avatar
Joseph Arceneaux committed
1
/* Output like sprintf to a buffer of specified size.
2 3
   Also takes args differently: pass one pointer to the end
   of the format string in addition to the format string itself.
4
   Copyright (C) 1985, 2001-2011  Free Software Foundation, Inc.
Joseph Arceneaux's avatar
Joseph Arceneaux committed
5 6 7

This file is part of GNU Emacs.

8
GNU Emacs is free software: you can redistribute it and/or modify
Joseph Arceneaux's avatar
Joseph Arceneaux committed
9
it under the terms of the GNU General Public License as published by
10 11
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Joseph Arceneaux's avatar
Joseph Arceneaux committed
12 13 14 15 16 17 18

GNU Emacs 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
19
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Joseph Arceneaux's avatar
Joseph Arceneaux committed
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
/* If you think about replacing this with some similar standard C function of
   the printf family (such as vsnprintf), please note that this function
   supports the following Emacs-specific features:

   . For %c conversions, it produces a string with the multibyte representation
     of the (`int') argument, suitable for display in an Emacs buffer.

   . For %s and %c, when field width is specified (e.g., %25s), it accounts for
     the diplay width of each character, according to char-width-table.  That
     is, it does not assume that each character takes one column on display.

   . If the size of the buffer is not enough to produce the formatted string in
     its entirety, it makes sure that truncation does not chop the last
     character in the middle of its multibyte sequence, producing an invalid
     sequence.

   . It accepts a pointer to the end of the format string, so the format string
     could include embedded null characters.

   . It signals an error if the length of the formatted string is about to
     overflow MOST_POSITIVE_FIXNUM, to avoid producing strings longer than what
     Emacs can handle.

   OTOH, this function supports only a small subset of the standard C formatted
   output facilities.  E.g., %u and %ll are not supported, and precision is
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
   ignored %s and %c conversions.  (See below for the detailed documentation of
   what is supported.)  However, this is okay, as this function is supposed to
   be called from `error' and similar functions, and thus does not need to
   support features beyond those in `Fformat', which is used by `error' on the
   Lisp level.  */

/* This function supports the following %-sequences in the `format'
   argument:

   %s means print a string argument.
   %S is silently treated as %s, for loose compatibility with `Fformat'.
   %d means print a `signed int' argument in decimal.
   %o means print an `unsigned int' argument in octal.
   %x means print an `unsigned int' argument in hex.
   %e means print a `double' argument in exponential notation.
   %f means print a `double' argument in decimal-point notation.
   %g means print a `double' argument in exponential notation
      or in decimal-point notation, whichever uses fewer characters.
   %c means print a `signed int' argument as a single character.
   %% means produce a literal % character.

67 68
   A %-sequence may contain optional flag, width, and precision specifiers, and
   a length modifier, as follows:
69

70
     %<flags><width><precision><length>character
71

Paul Eggert's avatar
Paul Eggert committed
72
   where flags is [+ -0], width is [0-9]+, precision is .[0-9]+, and length
73 74 75
   is empty or l or the value of the pI macro.  Also, %% in a format
   stands for a single % in the output.  A % that does not introduce a
   valid %-sequence causes undefined behavior.
76 77

   The + flag character inserts a + before any positive number, while a space
78 79 80 81 82 83 84 85
   inserts a space before any positive number; these flags only affect %d, %o,
   %x, %e, %f, and %g sequences.  The - and 0 flags affect the width specifier,
   as described below.  For signed numerical arguments only, the ` ' (space)
   flag causes the result to be prefixed with a space character if it does not
   start with a sign (+ or -).

   The l (lower-case letter ell) length modifier is a `long' data type
   modifier: it is supported for %d, %o, and %x conversions of integral
Paul Eggert's avatar
Paul Eggert committed
86
   arguments, must immediately precede the conversion specifier, and means that
87
   the respective argument is to be treated as `long int' or `unsigned long
88 89
   int'.  Similarly, the value of the pI macro means to use EMACS_INT or
   EMACS_UINT and the empty length modifier means `int' or `unsigned int'.
90 91 92 93 94 95 96 97 98 99

   The width specifier supplies a lower limit for the length of the printed
   representation.  The padding, if any, normally goes on the left, but it goes
   on the right if the - flag is present.  The padding character is normally a
   space, but (for numerical arguments only) it is 0 if the 0 flag is present.
   The - flag takes precedence over the 0 flag.

   For %e, %f, and %g sequences, the number after the "." in the precision
   specifier says how many decimal places to show; if zero, the decimal point
   itself is omitted.  For %s and %S, the precision specifier is ignored.  */
Joseph Arceneaux's avatar
Joseph Arceneaux committed
100

Richard M. Stallman's avatar
Richard M. Stallman committed
101
#include <config.h>
Joseph Arceneaux's avatar
Joseph Arceneaux committed
102 103
#include <stdio.h>
#include <ctype.h>
104
#include <setjmp.h>
105
#include <float.h>
Dave Love's avatar
Dave Love committed
106
#include <unistd.h>
107 108
#include <limits.h>

109 110
#include "lisp.h"

Karl Heuer's avatar
Karl Heuer committed
111 112 113
/* Since we use the macro CHAR_HEAD_P, we have to include this, but
   don't have to include others because CHAR_HEAD_P does not contains
   another macro.  */
114
#include "character.h"
Karl Heuer's avatar
Karl Heuer committed
115

116 117 118 119
#ifndef DBL_MAX_10_EXP
#define DBL_MAX_10_EXP 308 /* IEEE double */
#endif

120 121
/* Generate output from a format-spec FORMAT,
   terminated at position FORMAT_END.
122
   (*FORMAT_END is not part of the format, but must exist and be readable.)
123
   Output goes in BUFFER, which has room for BUFSIZE chars.
124 125 126
   BUFSIZE must be positive.  If the output does not fit, truncate it
   to fit and return BUFSIZE - 1; if this truncates a multibyte
   sequence, store '\0' into the sequence's first byte.
127 128
   Returns the number of bytes stored into BUFFER, excluding
   the terminating null byte.  Output is always null-terminated.
Richard M. Stallman's avatar
Richard M. Stallman committed
129 130
   String arguments are passed as C strings.
   Integers are passed as C integers.  */
131

132 133
size_t
doprnt (char *buffer, register size_t bufsize, const char *format,
134
	const char *format_end, va_list ap)
Joseph Arceneaux's avatar
Joseph Arceneaux committed
135
{
136
  const char *fmt = format;	/* Pointer into format string */
Joseph Arceneaux's avatar
Joseph Arceneaux committed
137
  register char *bufptr = buffer; /* Pointer into output buffer.. */
138

139
  /* Use this for sprintf unless we need something really big.  */
140
  char tembuf[DBL_MAX_10_EXP + 100];
141

142
  /* Size of sprintf_buffer.  */
143
  size_t size_allocated = sizeof (tembuf);
144

145 146
  /* Buffer to use for sprintf.  Either tembuf or same as BIG_BUFFER.  */
  char *sprintf_buffer = tembuf;
147

148
  /* Buffer we have got with malloc.  */
149
  char *big_buffer = NULL;
150

151
  register size_t tem;
152
  char *string;
153 154
  char fixed_buffer[20];	/* Default buffer for small formatting. */
  char *fmtcpy;
Joseph Arceneaux's avatar
Joseph Arceneaux committed
155
  int minlen;
156
  char charbuf[MAX_MULTIBYTE_LENGTH + 1];	/* Used for %c.  */
157
  USE_SAFE_ALLOCA;
Joseph Arceneaux's avatar
Joseph Arceneaux committed
158 159 160 161

  if (format_end == 0)
    format_end = format + strlen (format);

162 163 164
  if ((format_end - format + 1) < sizeof (fixed_buffer))
    fmtcpy = fixed_buffer;
  else
165
    SAFE_ALLOCA (fmtcpy, char *, format_end - format + 1);
166

Joseph Arceneaux's avatar
Joseph Arceneaux committed
167
  bufsize--;
168 169

  /* Loop until end of format string or buffer full. */
170
  while (fmt < format_end && bufsize > 0)
Joseph Arceneaux's avatar
Joseph Arceneaux committed
171 172 173
    {
      if (*fmt == '%')	/* Check for a '%' character */
	{
174 175 176
	  size_t size_bound = 0;
	  EMACS_INT width;  /* Columns occupied by STRING on display.  */
	  int long_flag = 0;
177
	  int pIlen = sizeof pI - 1;
178

Joseph Arceneaux's avatar
Joseph Arceneaux committed
179
	  fmt++;
Jim Blandy's avatar
Jim Blandy committed
180
	  /* Copy this one %-spec into fmtcpy.  */
181
	  string = fmtcpy;
Joseph Arceneaux's avatar
Joseph Arceneaux committed
182
	  *string++ = '%';
183
	  while (fmt < format_end)
Joseph Arceneaux's avatar
Joseph Arceneaux committed
184 185
	    {
	      *string++ = *fmt;
186 187 188 189 190 191
	      if ('0' <= *fmt && *fmt <= '9')
		{
		  /* Get an idea of how much space we might need.
		     This might be a field width or a precision; e.g.
		     %1.1000f and %1000.1f both might need 1000+ bytes.
		     Parse the width or precision, checking for overflow.  */
192
		  size_t n = *fmt - '0';
193
		  while (fmt + 1 < format_end
194
			 && '0' <= fmt[1] && fmt[1] <= '9')
195
		    {
Paul Eggert's avatar
Paul Eggert committed
196 197 198 199 200
		      /* Avoid size_t overflow.  Avoid int overflow too, as
			 many sprintfs mishandle widths greater than INT_MAX.
			 This test is simple but slightly conservative: e.g.,
			 (INT_MAX - INT_MAX % 10) is reported as an overflow
			 even when it's not.  */
201
		      if (n >= min (INT_MAX, SIZE_MAX) / 10)
202
			error ("Format width or precision too large");
Andreas Schwab's avatar
Andreas Schwab committed
203
		      n = n * 10 + fmt[1] - '0';
204 205 206 207 208 209
		      *string++ = *++fmt;
		    }

		  if (size_bound < n)
		    size_bound = n;
		}
210 211
	      else if (! (*fmt == '-' || *fmt == ' ' || *fmt == '.'
			  || *fmt == '+'))
Joseph Arceneaux's avatar
Joseph Arceneaux committed
212 213 214
		break;
	      fmt++;
	    }
215 216 217 218 219 220 221 222 223 224 225 226 227 228

	  if (0 < pIlen && pIlen <= format_end - fmt
	      && memcmp (fmt, pI, pIlen) == 0)
	    {
	      long_flag = 2;
	      memcpy (string, fmt + 1, pIlen);
	      string += pIlen;
	      fmt += pIlen;
	    }
	  else if (fmt < format_end && *fmt == 'l')
	    {
	      long_flag = 1;
	      *string++ = *++fmt;
	    }
Joseph Arceneaux's avatar
Joseph Arceneaux committed
229
	  *string = 0;
230

231 232
	  /* Make the size bound large enough to handle floating point formats
	     with large numbers.  */
233
	  if (size_bound > SIZE_MAX - DBL_MAX_10_EXP - 50)
234
	    error ("Format width or precision too large");
235
	  size_bound += DBL_MAX_10_EXP + 50;
236

237 238 239 240
	  /* Make sure we have that much.  */
	  if (size_bound > size_allocated)
	    {
	      if (big_buffer)
241 242
		xfree (big_buffer);
	      big_buffer = (char *) xmalloc (size_bound);
243 244 245
	      sprintf_buffer = big_buffer;
	      size_allocated = size_bound;
	    }
Joseph Arceneaux's avatar
Joseph Arceneaux committed
246 247 248 249
	  minlen = 0;
	  switch (*fmt++)
	    {
	    default:
250
	      error ("Invalid format operation %s", fmtcpy);
Joseph Arceneaux's avatar
Joseph Arceneaux committed
251 252

/*	    case 'b': */
253
	    case 'l':
Joseph Arceneaux's avatar
Joseph Arceneaux committed
254
	    case 'd':
255 256 257 258
	      {
		int i;
		long l;

259 260
		if (1 < long_flag)
		  {
261
		    EMACS_INT ll = va_arg (ap, EMACS_INT);
262 263 264
		    sprintf (sprintf_buffer, fmtcpy, ll);
		  }
		else if (long_flag)
265 266 267 268 269 270 271 272 273 274 275 276 277 278
		  {
		    l = va_arg(ap, long);
		    sprintf (sprintf_buffer, fmtcpy, l);
		  }
		else
		  {
		    i = va_arg(ap, int);
		    sprintf (sprintf_buffer, fmtcpy, i);
		  }
		/* Now copy into final output, truncating as necessary.  */
		string = sprintf_buffer;
		goto doit;
	      }

Joseph Arceneaux's avatar
Joseph Arceneaux committed
279 280
	    case 'o':
	    case 'x':
281 282 283 284
	      {
		unsigned u;
		unsigned long ul;

285 286
		if (1 < long_flag)
		  {
287
		    EMACS_UINT ull = va_arg (ap, EMACS_UINT);
288 289 290
		    sprintf (sprintf_buffer, fmtcpy, ull);
		  }
		else if (long_flag)
291 292 293 294 295 296 297 298 299 300 301 302 303
		  {
		    ul = va_arg(ap, unsigned long);
		    sprintf (sprintf_buffer, fmtcpy, ul);
		  }
		else
		  {
		    u = va_arg(ap, unsigned);
		    sprintf (sprintf_buffer, fmtcpy, u);
		  }
		/* Now copy into final output, truncating as necessary.  */
		string = sprintf_buffer;
		goto doit;
	      }
Joseph Arceneaux's avatar
Joseph Arceneaux committed
304

305 306 307 308
	    case 'f':
	    case 'e':
	    case 'g':
	      {
309 310
		double d = va_arg(ap, double);
		sprintf (sprintf_buffer, fmtcpy, d);
311
		/* Now copy into final output, truncating as necessary.  */
312
		string = sprintf_buffer;
313 314 315
		goto doit;
	      }

Joseph Arceneaux's avatar
Joseph Arceneaux committed
316 317 318 319 320
	    case 'S':
	      string[-1] = 's';
	    case 's':
	      if (fmtcpy[1] != 's')
		minlen = atoi (&fmtcpy[1]);
321
	      string = va_arg (ap, char *);
322
	      tem = strlen (string);
323
	      if (STRING_BYTES_BOUND < tem)
324
		error ("String for %%s or %%S format is too long");
Karl Heuer's avatar
Karl Heuer committed
325
	      width = strwidth (string, tem);
Richard M. Stallman's avatar
Richard M. Stallman committed
326 327
	      goto doit1;

Joseph Arceneaux's avatar
Joseph Arceneaux committed
328 329
	      /* Copy string into final output, truncating if no room.  */
	    doit:
Karl Heuer's avatar
Karl Heuer committed
330
	      /* Coming here means STRING contains ASCII only.  */
331
	      tem = strlen (string);
332
	      if (STRING_BYTES_BOUND < tem)
333 334
		error ("Format width or precision too large");
	      width = tem;
335
	    doit1:
Karl Heuer's avatar
Karl Heuer committed
336 337 338 339
	      /* We have already calculated:
		 TEM -- length of STRING,
		 WIDTH -- columns occupied by STRING when displayed, and
		 MINLEN -- minimum columns of the output.  */
Joseph Arceneaux's avatar
Joseph Arceneaux committed
340 341
	      if (minlen > 0)
		{
Karl Heuer's avatar
Karl Heuer committed
342
		  while (minlen > width && bufsize > 0)
Joseph Arceneaux's avatar
Joseph Arceneaux committed
343 344 345 346 347 348 349 350
		    {
		      *bufptr++ = ' ';
		      bufsize--;
		      minlen--;
		    }
		  minlen = 0;
		}
	      if (tem > bufsize)
Karl Heuer's avatar
Karl Heuer committed
351 352 353
		{
		  /* Truncate the string at character boundary.  */
		  tem = bufsize;
354
		  while (!CHAR_HEAD_P (string[tem - 1])) tem--;
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
		  /* If the multibyte sequence of this character is
		     too long for the space we have left in the
		     buffer, truncate before it.  */
		  if (tem > 0
		      && BYTES_BY_CHAR_HEAD (string[tem - 1]) > bufsize)
		    tem--;
		  if (tem > 0)
		    memcpy (bufptr, string, tem);
		  bufptr[tem] = 0;
		  /* Trigger exit from the loop, but make sure we
		     return to the caller a value which will indicate
		     that the buffer was too small.  */
		  bufptr += bufsize;
		  bufsize = 0;
		  continue;
Karl Heuer's avatar
Karl Heuer committed
370 371
		}
	      else
372
		memcpy (bufptr, string, tem);
Joseph Arceneaux's avatar
Joseph Arceneaux committed
373 374 375 376
	      bufptr += tem;
	      bufsize -= tem;
	      if (minlen < 0)
		{
Karl Heuer's avatar
Karl Heuer committed
377
		  while (minlen < - width && bufsize > 0)
Joseph Arceneaux's avatar
Joseph Arceneaux committed
378 379 380 381 382 383 384 385 386 387
		    {
		      *bufptr++ = ' ';
		      bufsize--;
		      minlen++;
		    }
		  minlen = 0;
		}
	      continue;

	    case 'c':
388
	      {
389 390
		int chr = va_arg(ap, int);
		tem = CHAR_STRING (chr, (unsigned char *) charbuf);
391 392 393 394 395 396 397
		string = charbuf;
		string[tem] = 0;
		width = strwidth (string, tem);
		if (fmtcpy[1] != 'c')
		  minlen = atoi (&fmtcpy[1]);
		goto doit1;
	      }
Joseph Arceneaux's avatar
Joseph Arceneaux committed
398 399 400 401 402

	    case '%':
	      fmt--;    /* Drop thru and this % will be treated as normal */
	    }
	}
Karl Heuer's avatar
Karl Heuer committed
403 404 405 406 407 408 409

      {
	/* Just some character; Copy it if the whole multi-byte form
	   fit in the buffer.  */
	char *save_bufptr = bufptr;

	do { *bufptr++ = *fmt++; }
410
	while (fmt < format_end && --bufsize > 0 && !CHAR_HEAD_P (*fmt));
411
	if (!CHAR_HEAD_P (*fmt))
Karl Heuer's avatar
Karl Heuer committed
412
	  {
413 414 415
	    /* Truncate, but return value that will signal to caller
	       that the buffer was too small.  */
	    *save_bufptr = 0;
Karl Heuer's avatar
Karl Heuer committed
416 417 418
	    break;
	  }
      }
Joseph Arceneaux's avatar
Joseph Arceneaux committed
419 420
    };

421
  /* If we had to malloc something, free it.  */
422
  xfree (big_buffer);
423

424
  *bufptr = 0;		/* Make sure our string ends with a '\0' */
425 426

  SAFE_FREE ();
Joseph Arceneaux's avatar
Joseph Arceneaux committed
427 428
  return bufptr - buffer;
}