lread.c 130 KB
Newer Older
Jim Blandy's avatar
Jim Blandy committed
1
/* Lisp parsing and input streams.
2 3 4 5

Copyright (C) 1985, 1986, 1987, 1988, 1989, 1993, 1994, 1995, 1997,
  1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008,
  2009, 2010  Free Software Foundation, Inc.
Jim Blandy's avatar
Jim Blandy committed
6 7 8

This file is part of GNU Emacs.

9
GNU Emacs is free software: you can redistribute it and/or modify
Jim Blandy's avatar
Jim Blandy committed
10
it under the terms of the GNU General Public License as published by
11 12
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
Jim Blandy's avatar
Jim Blandy committed
13 14 15 16 17 18 19

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
20
along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.  */
Jim Blandy's avatar
Jim Blandy committed
21 22


Richard M. Stallman's avatar
Richard M. Stallman committed
23
#include <config.h>
Jim Blandy's avatar
Jim Blandy committed
24 25 26 27
#include <stdio.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/file.h>
28
#include <errno.h>
29
#include <setjmp.h>
Jim Blandy's avatar
Jim Blandy committed
30
#include "lisp.h"
31
#include "intervals.h"
Jim Blandy's avatar
Jim Blandy committed
32
#include "buffer.h"
Kenichi Handa's avatar
Kenichi Handa committed
33
#include "character.h"
Karl Heuer's avatar
Karl Heuer committed
34
#include "charset.h"
Kenichi Handa's avatar
Kenichi Handa committed
35
#include "coding.h"
36
#include <epaths.h>
Jim Blandy's avatar
Jim Blandy committed
37
#include "commands.h"
Jim Blandy's avatar
Jim Blandy committed
38
#include "keyboard.h"
39
#include "frame.h"
40
#include "termhooks.h"
41
#include "coding.h"
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
42
#include "blockinput.h"
Jim Blandy's avatar
Jim Blandy committed
43

44 45 46 47
#ifdef MSDOS
#include "msdos.h"
#endif

Andreas Schwab's avatar
Andreas Schwab committed
48 49 50 51
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif

Jim Blandy's avatar
Jim Blandy committed
52 53
#include <math.h>

Richard M. Stallman's avatar
Richard M. Stallman committed
54 55 56 57
#ifdef HAVE_SETLOCALE
#include <locale.h>
#endif /* HAVE_SETLOCALE */

Dave Love's avatar
Dave Love committed
58
#include <fcntl.h>
Richard M. Stallman's avatar
Richard M. Stallman committed
59

60
#ifdef HAVE_FSEEKO
61 62 63 64 65 66 67
#define file_offset off_t
#define file_tell ftello
#else
#define file_offset long
#define file_tell ftell
#endif

68 69 70 71 72 73 74
/* hash table read constants */
Lisp_Object Qhash_table, Qdata;
Lisp_Object Qtest, Qsize;
Lisp_Object Qweakness;
Lisp_Object Qrehash_size;
Lisp_Object Qrehash_threshold;

75
Lisp_Object Qread_char, Qget_file_char, Qstandard_input, Qcurrent_load_list;
Jim Blandy's avatar
Jim Blandy committed
76
Lisp_Object Qvariable_documentation, Vvalues, Vstandard_input, Vafter_load_alist;
77
Lisp_Object Qascii_character, Qload, Qload_file_name;
78
Lisp_Object Qbackquote, Qcomma, Qcomma_at, Qcomma_dot, Qfunction;
79
Lisp_Object Qinhibit_file_name_operation;
80
Lisp_Object Qeval_buffer_list, Veval_buffer_list;
81
Lisp_Object Qlexical_binding;
82
Lisp_Object Qfile_truename, Qdo_after_load_evaluation; /* ACM 2006/5/16 */
83

Kenichi Handa's avatar
Kenichi Handa committed
84 85 86 87 88 89
/* Used instead of Qget_file_char while loading *.elc files compiled
   by Emacs 21 or older.  */
static Lisp_Object Qget_emacs_mule_file_char;

static Lisp_Object Qload_force_doc_strings;

90
extern Lisp_Object Qinternal_interpreter_environment;
Jim Blandy's avatar
Jim Blandy committed
91

Glenn Morris's avatar
Glenn Morris committed
92
/* non-zero if inside `load' */
Jim Blandy's avatar
Jim Blandy committed
93
int load_in_progress;
94
static Lisp_Object Qload_in_progress;
Ken Raeburn's avatar
Ken Raeburn committed
95

96 97 98
/* Directory in which the sources were found.  */
Lisp_Object Vsource_directory;

99
/* Search path and suffixes for files to be loaded. */
100
Lisp_Object Vload_path, Vload_suffixes, Vload_file_rep_suffixes;
Jim Blandy's avatar
Jim Blandy committed
101

Karl Heuer's avatar
Karl Heuer committed
102 103 104
/* File name of user's init file.  */
Lisp_Object Vuser_init_file;

105 106 107 108
/* This is the user-visible association list that maps features to
   lists of defs in their load files. */
Lisp_Object Vload_history;

109
/* This is used to build the load history. */
110 111
Lisp_Object Vcurrent_load_list;

112 113 114
/* List of files that were preloaded.  */
Lisp_Object Vpreloaded_file_list;

115 116 117
/* Name of file actually being read by `load'.  */
Lisp_Object Vload_file_name;

118 119 120
/* Function to use for reading, in `load' and friends.  */
Lisp_Object Vload_read_function;

121 122 123
/* Non-nil means read recursive structures using #n= and #n# syntax.  */
Lisp_Object Vread_circle;

124 125 126 127 128 129
/* The association list of objects read with the #n=object form.
   Each member of the list has the form (n . object), and is used to
   look up the object for the corresponding #n# construct.
   It must be set to nil before all top-level calls to read0.  */
Lisp_Object read_objects;

130 131 132
/* Nonzero means load should forcibly load all dynamic doc strings.  */
static int load_force_doc_strings;

133 134 135
/* Nonzero means read should convert strings to unibyte.  */
static int load_convert_to_unibyte;

Kenichi Handa's avatar
Kenichi Handa committed
136 137 138 139 140
/* Nonzero means READCHAR should read bytes one by one (not character)
   when READCHARFUN is Qget_file_char or Qget_emacs_mule_file_char.
   This is set to 1 by read1 temporarily while handling #@NUMBER.  */
static int load_each_byte;

141
/* Function to use for loading an Emacs Lisp source file (not
Karl Heuer's avatar
Karl Heuer committed
142 143 144
   compiled) instead of readevalloop.  */
Lisp_Object Vload_source_file_function;

145 146 147
/* List of all DEFVAR_BOOL variables.  Used by the byte optimizer.  */
Lisp_Object Vbyte_boolean_vars;

148 149 150 151 152 153 154
/* Whether or not to add a `read-positions' property to symbols
   read. */
Lisp_Object Vread_with_symbol_positions;

/* List of (SYMBOL . POSITION) accumulated so far. */
Lisp_Object Vread_symbol_positions_list;

155 156 157
/* If non-nil `readevalloop' evaluates code in a lexical environment.  */
Lisp_Object Vlexical_binding;

158 159 160
/* List of descriptors now open for Fload.  */
static Lisp_Object load_descriptor_list;

161
/* File for get_file_char to read from.  Use by load.  */
Jim Blandy's avatar
Jim Blandy committed
162 163 164 165 166
static FILE *instream;

/* When nonzero, read conses in pure space */
static int read_pure;

167
/* For use within read-from-string (this reader is non-reentrant!!)  */
168 169 170
static EMACS_INT read_from_string_index;
static EMACS_INT read_from_string_index_byte;
static EMACS_INT read_from_string_limit;
171

172 173
/* Number of characters read in the current call to Fread or
   Fread_from_string. */
174
static EMACS_INT readchar_count;
175

176
/* This contains the last string skipped with #@.  */
177 178 179 180 181 182
static char *saved_doc_string;
/* Length of buffer allocated in saved_doc_string.  */
static int saved_doc_string_size;
/* Length of actual data in saved_doc_string.  */
static int saved_doc_string_length;
/* This is the file position that string came from.  */
183
static file_offset saved_doc_string_position;
184

185 186 187 188 189 190 191 192 193
/* This contains the previous string skipped with #@.
   We copy it from saved_doc_string when a new string
   is put in saved_doc_string.  */
static char *prev_saved_doc_string;
/* Length of buffer allocated in prev_saved_doc_string.  */
static int prev_saved_doc_string_size;
/* Length of actual data in prev_saved_doc_string.  */
static int prev_saved_doc_string_length;
/* This is the file position that string came from.  */
194
static file_offset prev_saved_doc_string_position;
195

196 197 198 199 200
/* Nonzero means inside a new-style backquote
   with no surrounding parentheses.
   Fread initializes this to zero, so we need not specbind it
   or worry about what happens to it when there is an error.  */
static int new_backquote_flag;
201
static Lisp_Object Vold_style_backquotes, Qold_style_backquotes;
202 203 204 205

/* A list of file names for files being loaded in Fload.  Used to
   check for recursive loads.  */

Lute Kamstra's avatar
Lute Kamstra committed
206
static Lisp_Object Vloads_in_progress;
207

208 209 210 211
/* Non-zero means load dangerous compiled Lisp files.  */

int load_dangerous_libraries;

212 213 214 215
/* Non-zero means force printing messages when loading Lisp files.  */

int force_load_messages;

216 217 218 219
/* A regular expression used to detect files compiled with Emacs.  */

static Lisp_Object Vbytecomp_version_regexp;

220 221
static int read_emacs_mule_char (int, int (*) (int, Lisp_Object),
                                 Lisp_Object);
Kenichi Handa's avatar
Kenichi Handa committed
222

223
static void readevalloop (Lisp_Object, FILE*, Lisp_Object,
224
                          Lisp_Object (*) (Lisp_Object), int,
225 226 227 228
                          Lisp_Object, Lisp_Object,
                          Lisp_Object, Lisp_Object);
static Lisp_Object load_unwind (Lisp_Object);
static Lisp_Object load_descriptor_unwind (Lisp_Object);
229

230
static void invalid_syntax (const char *, int) NO_RETURN;
231
static void end_of_file_error (void) NO_RETURN;
232

Jim Blandy's avatar
Jim Blandy committed
233

Kenichi Handa's avatar
Kenichi Handa committed
234 235 236 237 238 239
/* Functions that read one byte from the current source READCHARFUN
   or unreads one byte.  If the integer argument C is -1, it returns
   one read byte, or -1 when there's no more byte in the source.  If C
   is 0 or positive, it unreads C, and the return value is not
   interesting.  */

240 241 242
static int readbyte_for_lambda (int, Lisp_Object);
static int readbyte_from_file (int, Lisp_Object);
static int readbyte_from_string (int, Lisp_Object);
Kenichi Handa's avatar
Kenichi Handa committed
243

Jim Blandy's avatar
Jim Blandy committed
244 245
/* Handle unreading and rereading of characters.
   Write READCHAR to read a character,
Karl Heuer's avatar
Karl Heuer committed
246 247
   UNREAD(c) to unread c to be read again.

Kenichi Handa's avatar
Kenichi Handa committed
248
   These macros correctly read/unread multibyte characters.  */
Jim Blandy's avatar
Jim Blandy committed
249

250
#define READCHAR readchar (readcharfun, NULL)
Jim Blandy's avatar
Jim Blandy committed
251 252
#define UNREAD(c) unreadchar (readcharfun, c)

253 254 255
/* Same as READCHAR but set *MULTIBYTE to the multibyteness of the source.  */
#define READCHAR_REPORT_MULTIBYTE(multibyte) readchar (readcharfun, multibyte)

Kenichi Handa's avatar
Kenichi Handa committed
256
/* When READCHARFUN is Qget_file_char, Qget_emacs_mule_file_char,
Kenichi Handa's avatar
Kenichi Handa committed
257 258
   Qlambda, or a cons, we use this to keep an unread character because
   a file stream can't handle multibyte-char unreading.  The value -1
Kenichi Handa's avatar
Kenichi Handa committed
259 260 261
   means that there's no unread character. */
static int unread_char;

Jim Blandy's avatar
Jim Blandy committed
262
static int
263
readchar (Lisp_Object readcharfun, int *multibyte)
Jim Blandy's avatar
Jim Blandy committed
264 265
{
  Lisp_Object tem;
266
  register int c;
267
  int (*readbyte) (int, Lisp_Object);
Kenichi Handa's avatar
Kenichi Handa committed
268 269 270
  unsigned char buf[MAX_MULTIBYTE_LENGTH];
  int i, len;
  int emacs_mule_encoding = 0;
Jim Blandy's avatar
Jim Blandy committed
271

272 273 274
  if (multibyte)
    *multibyte = 0;

275
  readchar_count++;
276

277
  if (BUFFERP (readcharfun))
Jim Blandy's avatar
Jim Blandy committed
278
    {
279
      register struct buffer *inbuffer = XBUFFER (readcharfun);
Jim Blandy's avatar
Jim Blandy committed
280

281
      EMACS_INT pt_byte = BUF_PT_BYTE (inbuffer);
282

283 284
      if (pt_byte >= BUF_ZV_BYTE (inbuffer))
	return -1;
Jim Blandy's avatar
Jim Blandy committed
285

286 287
      if (! NILP (inbuffer->enable_multibyte_characters))
	{
288
	  /* Fetch the character code from the buffer.  */
289 290
	  unsigned char *p = BUF_BYTE_ADDRESS (inbuffer, pt_byte);
	  BUF_INC_POS (inbuffer, pt_byte);
291
	  c = STRING_CHAR (p);
292 293
	  if (multibyte)
	    *multibyte = 1;
294 295 296 297
	}
      else
	{
	  c = BUF_FETCH_BYTE (inbuffer, pt_byte);
Kenichi Handa's avatar
Kenichi Handa committed
298 299
	  if (! ASCII_BYTE_P (c))
	    c = BYTE8_TO_CHAR (c);
300
	  pt_byte++;
301
	}
302
      SET_BUF_PT_BOTH (inbuffer, BUF_PT (inbuffer) + 1, pt_byte);
303

304
      return c;
Jim Blandy's avatar
Jim Blandy committed
305
    }
306
  if (MARKERP (readcharfun))
Jim Blandy's avatar
Jim Blandy committed
307
    {
308
      register struct buffer *inbuffer = XMARKER (readcharfun)->buffer;
Jim Blandy's avatar
Jim Blandy committed
309

310
      EMACS_INT bytepos = marker_byte_position (readcharfun);
311

312 313
      if (bytepos >= BUF_ZV_BYTE (inbuffer))
	return -1;
314

315 316
      if (! NILP (inbuffer->enable_multibyte_characters))
	{
317
	  /* Fetch the character code from the buffer.  */
318 319
	  unsigned char *p = BUF_BYTE_ADDRESS (inbuffer, bytepos);
	  BUF_INC_POS (inbuffer, bytepos);
320
	  c = STRING_CHAR (p);
321 322
	  if (multibyte)
	    *multibyte = 1;
323 324 325 326
	}
      else
	{
	  c = BUF_FETCH_BYTE (inbuffer, bytepos);
Kenichi Handa's avatar
Kenichi Handa committed
327 328
	  if (! ASCII_BYTE_P (c))
	    c = BYTE8_TO_CHAR (c);
329
	  bytepos++;
330 331
	}

332 333 334 335
      XMARKER (readcharfun)->bytepos = bytepos;
      XMARKER (readcharfun)->charpos++;

      return c;
Jim Blandy's avatar
Jim Blandy committed
336
    }
337 338

  if (EQ (readcharfun, Qlambda))
Kenichi Handa's avatar
Kenichi Handa committed
339 340 341 342
    {
      readbyte = readbyte_for_lambda;
      goto read_multibyte;
    }
343

Jim Blandy's avatar
Jim Blandy committed
344
  if (EQ (readcharfun, Qget_file_char))
345
    {
Kenichi Handa's avatar
Kenichi Handa committed
346 347
      readbyte = readbyte_from_file;
      goto read_multibyte;
348
    }
Jim Blandy's avatar
Jim Blandy committed
349

350
  if (STRINGP (readcharfun))
Jim Blandy's avatar
Jim Blandy committed
351
    {
352
      if (read_from_string_index >= read_from_string_limit)
Jim Blandy's avatar
Jim Blandy committed
353
	c = -1;
354 355 356 357 358 359 360 361
      else if (STRING_MULTIBYTE (readcharfun))
	{
	  if (multibyte)
	    *multibyte = 1;
	  FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c, readcharfun,
					      read_from_string_index,
					      read_from_string_index_byte);
	}
362
      else
363 364 365 366 367
	{
	  c = SREF (readcharfun, read_from_string_index_byte);
	  read_from_string_index++;
	  read_from_string_index_byte++;
	}
Jim Blandy's avatar
Jim Blandy committed
368 369 370
      return c;
    }

Kenichi Handa's avatar
Kenichi Handa committed
371 372 373 374 375 376 377 378 379 380 381 382
  if (CONSP (readcharfun))
    {
      /* This is the case that read_vector is reading from a unibyte
	 string that contains a byte sequence previously skipped
	 because of #@NUMBER.  The car part of readcharfun is that
	 string, and the cdr part is a value of readcharfun given to
	 read_vector.  */
      readbyte = readbyte_from_string;
      if (EQ (XCDR (readcharfun), Qget_emacs_mule_file_char))
	emacs_mule_encoding = 1;
      goto read_multibyte;
    }
Kenichi Handa's avatar
Kenichi Handa committed
383

Kenichi Handa's avatar
Kenichi Handa committed
384 385 386 387 388 389
  if (EQ (readcharfun, Qget_emacs_mule_file_char))
    {
      readbyte = readbyte_from_file;
      emacs_mule_encoding = 1;
      goto read_multibyte;
    }
Jim Blandy's avatar
Jim Blandy committed
390 391 392

  tem = call0 (readcharfun);

Jim Blandy's avatar
Jim Blandy committed
393
  if (NILP (tem))
Jim Blandy's avatar
Jim Blandy committed
394 395
    return -1;
  return XINT (tem);
Kenichi Handa's avatar
Kenichi Handa committed
396 397 398 399 400 401 402 403 404

 read_multibyte:
  if (unread_char >= 0)
    {
      c = unread_char;
      unread_char = -1;
      return c;
    }
  c = (*readbyte) (-1, readcharfun);
405 406 407 408 409
  if (c < 0 || load_each_byte)
    return c;
  if (multibyte)
    *multibyte = 1;
  if (ASCII_BYTE_P (c))
Kenichi Handa's avatar
Kenichi Handa committed
410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426
    return c;
  if (emacs_mule_encoding)
    return read_emacs_mule_char (c, readbyte, readcharfun);
  i = 0;
  buf[i++] = c;
  len = BYTES_BY_CHAR_HEAD (c);
  while (i < len)
    {
      c = (*readbyte) (-1, readcharfun);
      if (c < 0 || ! TRAILING_CODE_P (c))
	{
	  while (--i > 1)
	    (*readbyte) (buf[i], readcharfun);
	  return BYTE8_TO_CHAR (buf[0]);
	}
      buf[i++] = c;
    }
427
  return STRING_CHAR (buf);
Jim Blandy's avatar
Jim Blandy committed
428 429 430 431 432 433
}

/* Unread the character C in the way appropriate for the stream READCHARFUN.
   If the stream is a user function, call it with the char as argument.  */

static void
434
unreadchar (Lisp_Object readcharfun, int c)
Jim Blandy's avatar
Jim Blandy committed
435
{
436
  readchar_count--;
437 438 439 440
  if (c == -1)
    /* Don't back up the pointer if we're unreading the end-of-input mark,
       since readchar didn't advance it when we read it.  */
    ;
441
  else if (BUFFERP (readcharfun))
442
    {
443
      struct buffer *b = XBUFFER (readcharfun);
444
      EMACS_INT bytepos = BUF_PT_BYTE (b);
445

Kenichi Handa's avatar
Kenichi Handa committed
446 447 448
      BUF_PT (b)--;
      if (! NILP (b->enable_multibyte_characters))
	BUF_DEC_POS (b, bytepos);
449
      else
Kenichi Handa's avatar
Kenichi Handa committed
450
	bytepos--;
451

Kenichi Handa's avatar
Kenichi Handa committed
452
      BUF_PT_BYTE (b) = bytepos;
453
    }
454
  else if (MARKERP (readcharfun))
455
    {
456
      struct buffer *b = XMARKER (readcharfun)->buffer;
457
      EMACS_INT bytepos = XMARKER (readcharfun)->bytepos;
458

Kenichi Handa's avatar
Kenichi Handa committed
459 460 461
      XMARKER (readcharfun)->charpos--;
      if (! NILP (b->enable_multibyte_characters))
	BUF_DEC_POS (b, bytepos);
462
      else
Kenichi Handa's avatar
Kenichi Handa committed
463
	bytepos--;
464

Kenichi Handa's avatar
Kenichi Handa committed
465
      XMARKER (readcharfun)->bytepos = bytepos;
466
    }
467
  else if (STRINGP (readcharfun))
468 469 470 471 472
    {
      read_from_string_index--;
      read_from_string_index_byte
	= string_char_to_byte (readcharfun, read_from_string_index);
    }
Kenichi Handa's avatar
Kenichi Handa committed
473 474 475 476
  else if (CONSP (readcharfun))
    {
      unread_char = c;
    }
477
  else if (EQ (readcharfun, Qlambda))
Kenichi Handa's avatar
Kenichi Handa committed
478 479 480 481 482 483 484
    {
      unread_char = c;
    }
  else if (EQ (readcharfun, Qget_file_char)
	   || EQ (readcharfun, Qget_emacs_mule_file_char))
    {
      if (load_each_byte)
Miles Bader's avatar
Miles Bader committed
485 486 487 488 489
	{
	  BLOCK_INPUT;
	  ungetc (c, instream);
	  UNBLOCK_INPUT;
	}
Kenichi Handa's avatar
Kenichi Handa committed
490 491 492
      else
	unread_char = c;
    }
Jim Blandy's avatar
Jim Blandy committed
493 494 495 496
  else
    call1 (readcharfun, make_number (c));
}

Kenichi Handa's avatar
Kenichi Handa committed
497
static int
498
readbyte_for_lambda (int c, Lisp_Object readcharfun)
Kenichi Handa's avatar
Kenichi Handa committed
499 500 501 502 503 504
{
  return read_bytecode_char (c >= 0);
}


static int
505
readbyte_from_file (int c, Lisp_Object readcharfun)
Kenichi Handa's avatar
Kenichi Handa committed
506 507 508
{
  if (c >= 0)
    {
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
509
      BLOCK_INPUT;
Kenichi Handa's avatar
Kenichi Handa committed
510
      ungetc (c, instream);
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
511
      UNBLOCK_INPUT;
Kenichi Handa's avatar
Kenichi Handa committed
512 513 514
      return 0;
    }

Miles Bader's avatar
Miles Bader committed
515
  BLOCK_INPUT;
Kenichi Handa's avatar
Kenichi Handa committed
516
  c = getc (instream);
Miles Bader's avatar
Miles Bader committed
517

Kenichi Handa's avatar
Kenichi Handa committed
518
#ifdef EINTR
Miles Bader's avatar
Miles Bader committed
519 520 521
  /* Interrupted reads have been observed while reading over the network */
  while (c == EOF && ferror (instream) && errno == EINTR)
    {
Miles Bader's avatar
Miles Bader committed
522
      UNBLOCK_INPUT;
Miles Bader's avatar
Miles Bader committed
523 524
      QUIT;
      BLOCK_INPUT;
Miles Bader's avatar
Miles Bader committed
525
      clearerr (instream);
Miles Bader's avatar
Miles Bader committed
526
      c = getc (instream);
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
527
    }
Kenichi Handa's avatar
Kenichi Handa committed
528
#endif
Miles Bader's avatar
Miles Bader committed
529

Miles Bader's avatar
Miles Bader committed
530 531
  UNBLOCK_INPUT;

Kenichi Handa's avatar
Kenichi Handa committed
532 533 534 535
  return (c == EOF ? -1 : c);
}

static int
536
readbyte_from_string (int c, Lisp_Object readcharfun)
Kenichi Handa's avatar
Kenichi Handa committed
537 538 539 540 541 542 543 544 545
{
  Lisp_Object string = XCAR (readcharfun);

  if (c >= 0)
    {
      read_from_string_index--;
      read_from_string_index_byte
	= string_char_to_byte (string, read_from_string_index);
    }
Kenichi Handa's avatar
Kenichi Handa committed
546

Kenichi Handa's avatar
Kenichi Handa committed
547 548 549 550 551 552 553 554 555 556 557 558 559 560 561
  if (read_from_string_index >= read_from_string_limit)
    c = -1;
  else
    FETCH_STRING_CHAR_ADVANCE (c, string,
			       read_from_string_index,
			       read_from_string_index_byte);
  return c;
}


/* Read one non-ASCII character from INSTREAM.  The character is
   encoded in `emacs-mule' and the first byte is already read in
   C.  */

static int
562
read_emacs_mule_char (int c, int (*readbyte) (int, Lisp_Object), Lisp_Object readcharfun)
Kenichi Handa's avatar
Kenichi Handa committed
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
{
  /* Emacs-mule coding uses at most 4-byte for one character.  */
  unsigned char buf[4];
  int len = emacs_mule_bytes[c];
  struct charset *charset;
  int i;
  unsigned code;

  if (len == 1)
    /* C is not a valid leading-code of `emacs-mule'.  */
    return BYTE8_TO_CHAR (c);

  i = 0;
  buf[i++] = c;
  while (i < len)
    {
      c = (*readbyte) (-1, readcharfun);
      if (c < 0xA0)
	{
	  while (--i > 1)
	    (*readbyte) (buf[i], readcharfun);
	  return BYTE8_TO_CHAR (buf[0]);
	}
      buf[i++] = c;
    }

  if (len == 2)
    {
591
      charset = CHARSET_FROM_ID (emacs_mule_charset[buf[0]]);
Kenichi Handa's avatar
Kenichi Handa committed
592 593 594 595 596 597 598
      code = buf[1] & 0x7F;
    }
  else if (len == 3)
    {
      if (buf[0] == EMACS_MULE_LEADING_CODE_PRIVATE_11
	  || buf[0] == EMACS_MULE_LEADING_CODE_PRIVATE_12)
	{
599
	  charset = CHARSET_FROM_ID (emacs_mule_charset[buf[1]]);
Kenichi Handa's avatar
Kenichi Handa committed
600 601 602 603
	  code = buf[2] & 0x7F;
	}
      else
	{
604
	  charset = CHARSET_FROM_ID (emacs_mule_charset[buf[0]]);
Kenichi Handa's avatar
Kenichi Handa committed
605 606 607 608 609
	  code = ((buf[1] << 8) | buf[2]) & 0x7F7F;
	}
    }
  else
    {
610
      charset = CHARSET_FROM_ID (emacs_mule_charset[buf[1]]);
Kenichi Handa's avatar
Kenichi Handa committed
611
      code = ((buf[2] << 8) | buf[3]) & 0x7F7F;
Kenichi Handa's avatar
Kenichi Handa committed
612 613 614 615 616 617 618 619 620
    }
  c = DECODE_CHAR (charset, code);
  if (c < 0)
    Fsignal (Qinvalid_read_syntax,
	     Fcons (build_string ("invalid multibyte form"), Qnil));
  return c;
}


621 622 623 624
static Lisp_Object read_internal_start (Lisp_Object, Lisp_Object,
                                        Lisp_Object);
static Lisp_Object read0 (Lisp_Object);
static Lisp_Object read1 (Lisp_Object, int *, int);
625

626 627
static Lisp_Object read_list (int, Lisp_Object);
static Lisp_Object read_vector (Lisp_Object, int);
628

629 630 631 632 633
static Lisp_Object substitute_object_recurse (Lisp_Object, Lisp_Object,
                                              Lisp_Object);
static void substitute_object_in_subtree (Lisp_Object,
                                          Lisp_Object);
static void substitute_in_interval (INTERVAL, Lisp_Object);
634

Jim Blandy's avatar
Jim Blandy committed
635

636
/* Get a character from the tty.  */
Jim Blandy's avatar
Jim Blandy committed
637

638 639 640 641 642 643 644 645 646 647 648 649 650
/* Read input events until we get one that's acceptable for our purposes.

   If NO_SWITCH_FRAME is non-zero, switch-frame events are stashed
   until we get a character we like, and then stuffed into
   unread_switch_frame.

   If ASCII_REQUIRED is non-zero, we check function key events to see
   if the unmodified version of the symbol has a Qascii_character
   property, and use that character, if present.

   If ERROR_NONASCII is non-zero, we signal an error if the input we
   get isn't an ASCII character with modifiers.  If it's zero but
   ASCII_REQUIRED is non-zero, we just re-read until we get an ASCII
651 652 653
   character.

   If INPUT_METHOD is nonzero, we invoke the current input method
654 655 656 657
   if the character warrants that.

   If SECONDS is a number, we wait that many seconds for input, and
   return Qnil if no input arrives within that time.  */
658

659
Lisp_Object
Andreas Schwab's avatar
Andreas Schwab committed
660 661
read_filtered_event (int no_switch_frame, int ascii_required,
		     int error_nonascii, int input_method, Lisp_Object seconds)
662
{
663
  Lisp_Object val, delayed_switch_frame;
664
  EMACS_TIME end_time;
665

666
#ifdef HAVE_WINDOW_SYSTEM
667 668
  if (display_hourglass_p)
    cancel_hourglass ();
669
#endif
670

671
  delayed_switch_frame = Qnil;
672

673 674 675 676 677
  /* Compute timeout.  */
  if (NUMBERP (seconds))
    {
      EMACS_TIME wait_time;
      int sec, usec;
678
      double duration = extract_float (seconds);
679 680 681 682 683 684 685 686

      sec  = (int) duration;
      usec = (duration - sec) * 1000000;
      EMACS_GET_TIME (end_time);
      EMACS_SET_SECS_USECS (wait_time, sec, usec);
      EMACS_ADD_TIME (end_time, end_time, wait_time);
    }

Karoly Lorentey's avatar
Karoly Lorentey committed
687
/* Read until we get an acceptable event.  */
688
 retry:
689
  do
Karoly Lorentey's avatar
Karoly Lorentey committed
690 691
    val = read_char (0, 0, 0, (input_method ? Qnil : Qt), 0,
		     NUMBERP (seconds) ? &end_time : NULL);
692
  while (INTEGERP (val) && XINT (val) == -2); /* wrong_kboard_jmpbuf */
693

694
  if (BUFFERP (val))
695 696
    goto retry;

697
  /* switch-frame events are put off until after the next ASCII
Karl Heuer's avatar
Karl Heuer committed
698
     character.  This is better than signaling an error just because
699 700 701 702 703
     the last characters were typed to a separate minibuffer frame,
     for example.  Eventually, some code which can deal with
     switch-frame events will read it and process it.  */
  if (no_switch_frame
      && EVENT_HAS_PARAMETERS (val)
704
      && EQ (EVENT_HEAD_KIND (EVENT_HEAD (val)), Qswitch_frame))
705 706 707 708 709
    {
      delayed_switch_frame = val;
      goto retry;
    }

710
  if (ascii_required && !(NUMBERP (seconds) && NILP (val)))
711 712
    {
      /* Convert certain symbols to their ASCII equivalents.  */
713
      if (SYMBOLP (val))
714
	{
715
	  Lisp_Object tem, tem1;
716 717 718 719 720 721 722
	  tem = Fget (val, Qevent_symbol_element_mask);
	  if (!NILP (tem))
	    {
	      tem1 = Fget (Fcar (tem), Qascii_character);
	      /* Merge this symbol's modifier bits
		 with the ASCII equivalent of its basic code.  */
	      if (!NILP (tem1))
723
		XSETFASTINT (val, XINT (tem1) | XINT (Fcar (Fcdr (tem))));
724 725
	    }
	}
726

727
      /* If we don't have a character now, deal with it appropriately.  */
728
      if (!INTEGERP (val))
729 730 731
	{
	  if (error_nonascii)
	    {
732
	      Vunread_command_events = Fcons (val, Qnil);
733 734 735 736 737 738 739 740 741 742
	      error ("Non-character input-event");
	    }
	  else
	    goto retry;
	}
    }

  if (! NILP (delayed_switch_frame))
    unread_switch_frame = delayed_switch_frame;

743 744
#if 0

745
#ifdef HAVE_WINDOW_SYSTEM
746 747
  if (display_hourglass_p)
    start_hourglass ();
748
#endif
749 750 751

#endif

752 753 754
  return val;
}

755
DEFUN ("read-char", Fread_char, Sread_char, 0, 3, 0,
756 757
       doc: /* Read a character from the command input (keyboard or macro).
It is returned as a number.
758 759 760
If the character has modifiers, they are resolved and reflected to the
character code if possible (e.g. C-SPC -> 0).

761 762
If the user generates an event which is not a character (i.e. a mouse
click or function key event), `read-char' signals an error.  As an
763 764
exception, switch-frame events are put off until non-character events
can be read.
765 766 767 768 769 770
If you want to read non-character events, or ignore them, call
`read-event' or `read-char-exclusive' instead.

If the optional argument PROMPT is non-nil, display that as a prompt.
If the optional argument INHERIT-INPUT-METHOD is non-nil and some
input method is turned on in the current buffer, that input method
771 772 773 774 775
is used for reading a character.
If the optional argument SECONDS is non-nil, it should be a number
specifying the maximum number of seconds to wait for input.  If no
input arrives in that time, return nil.  SECONDS may be a
floating-point value.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
776
  (Lisp_Object prompt, Lisp_Object inherit_input_method, Lisp_Object seconds)
Jim Blandy's avatar
Jim Blandy committed
777
{
778 779
  Lisp_Object val;

780 781
  if (! NILP (prompt))
    message_with_string ("%s", prompt, 0);
782
  val = read_filtered_event (1, 1, 1, ! NILP (inherit_input_method), seconds);
783 784 785

  return (NILP (val) ? Qnil
	  : make_number (char_resolve_modifier_mask (XINT (val))));
Jim Blandy's avatar
Jim Blandy committed
786 787
}

788
DEFUN ("read-event", Fread_event, Sread_event, 0, 3, 0,
789 790 791 792
       doc: /* Read an event object from the input stream.
If the optional argument PROMPT is non-nil, display that as a prompt.
If the optional argument INHERIT-INPUT-METHOD is non-nil and some
input method is turned on in the current buffer, that input method
793 794 795 796 797
is used for reading a character.
If the optional argument SECONDS is non-nil, it should be a number
specifying the maximum number of seconds to wait for input.  If no
input arrives in that time, return nil.  SECONDS may be a
floating-point value.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
798
  (Lisp_Object prompt, Lisp_Object inherit_input_method, Lisp_Object seconds)
Jim Blandy's avatar
Jim Blandy committed
799
{
800 801
  if (! NILP (prompt))
    message_with_string ("%s", prompt, 0);
802
  return read_filtered_event (0, 0, 0, ! NILP (inherit_input_method), seconds);
Jim Blandy's avatar
Jim Blandy committed
803 804
}

805
DEFUN ("read-char-exclusive", Fread_char_exclusive, Sread_char_exclusive, 0, 3, 0,
806 807
       doc: /* Read a character from the command input (keyboard or macro).
It is returned as a number.  Non-character events are ignored.
808 809
If the character has modifiers, they are resolved and reflected to the
character code if possible (e.g. C-SPC -> 0).
810 811 812 813

If the optional argument PROMPT is non-nil, display that as a prompt.
If the optional argument INHERIT-INPUT-METHOD is non-nil and some
input method is turned on in the current buffer, that input method
814 815 816 817 818
is used for reading a character.
If the optional argument SECONDS is non-nil, it should be a number
specifying the maximum number of seconds to wait for input.  If no
input arrives in that time, return nil.  SECONDS may be a
floating-point value.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
819
  (Lisp_Object prompt, Lisp_Object inherit_input_method, Lisp_Object seconds)
Jim Blandy's avatar
Jim Blandy committed
820
{
821 822
  Lisp_Object val;

823 824
  if (! NILP (prompt))
    message_with_string ("%s", prompt, 0);
825

826
  val = read_filtered_event (1, 1, 0, ! NILP (inherit_input_method), seconds);
827 828 829

  return (NILP (val) ? Qnil
	  : make_number (char_resolve_modifier_mask (XINT (val))));
Jim Blandy's avatar
Jim Blandy committed
830 831 832
}

DEFUN ("get-file-char", Fget_file_char, Sget_file_char, 0, 0, 0,
833
       doc: /* Don't use this yourself.  */)
Dan Nicolaescu's avatar
Dan Nicolaescu committed
834
  (void)
Jim Blandy's avatar
Jim Blandy committed
835 836
{
  register Lisp_Object val;
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
837
  BLOCK_INPUT;
838
  XSETINT (val, getc (instream));
YAMAMOTO Mitsuharu's avatar
YAMAMOTO Mitsuharu committed
839
  UNBLOCK_INPUT;
Jim Blandy's avatar
Jim Blandy committed
840 841
  return val;
}
842 843


844 845 846 847 848 849 850 851


/* Return true if the lisp code read using READCHARFUN defines a non-nil
   `lexical-binding' file variable.  After returning, the stream is
   positioned following the first line, if it is a comment, otherwise
   nothing is read.  */

static int
Stefan Monnier's avatar
Stefan Monnier committed
852
lisp_file_lexically_bound_p (Lisp_Object readcharfun)
853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954
{
  int ch = READCHAR;
  if (ch != ';')
    /* The first line isn't a comment, just give up.  */
    {
      UNREAD (ch);
      return 0;
    }
  else
    /* Look for an appropriate file-variable in the first line.  */
    {
      int rv = 0;
      enum {
	NOMINAL, AFTER_FIRST_DASH, AFTER_ASTERIX,
      } beg_end_state = NOMINAL;
      int in_file_vars = 0;

#define UPDATE_BEG_END_STATE(ch)					      \
  if (beg_end_state == NOMINAL)						      \
    beg_end_state = (ch == '-' ? AFTER_FIRST_DASH : NOMINAL);		      \
  else if (beg_end_state == AFTER_FIRST_DASH)				      \
    beg_end_state = (ch == '*' ? AFTER_ASTERIX : NOMINAL);		      \
  else if (beg_end_state == AFTER_ASTERIX)				      \
    {									      \
      if (ch == '-')							      \
	in_file_vars = !in_file_vars;					      \
      beg_end_state = NOMINAL;						      \
    }

      /* Skip until we get to the file vars, if any.  */
      do
	{
	  ch = READCHAR;
	  UPDATE_BEG_END_STATE (ch);
	}
      while (!in_file_vars && ch != '\n' && ch != EOF);

      while (in_file_vars)
	{
	  char var[100], *var_end, val[100], *val_end;

	  ch = READCHAR;

	  /* Read a variable name.  */
	  while (ch == ' ' || ch == '\t')
	    ch = READCHAR;

	  var_end = var;
	  while (ch != ':' && ch != '\n' && ch != EOF)
	    {
	      if (var_end < var + sizeof var - 1)
		*var_end++ = ch;
	      UPDATE_BEG_END_STATE (ch);
	      ch = READCHAR;
	    }
	  
	  while (var_end > var
		 && (var_end[-1] == ' ' || var_end[-1] == '\t'))
	    var_end--;
	  *var_end = '\0';

	  if (ch == ':')
	    {
	      /* Read a variable value.  */
	      ch = READCHAR;

	      while (ch == ' ' || ch == '\t')
		ch = READCHAR;

	      val_end = val;
	      while (ch != ';' && ch != '\n' && ch != EOF && in_file_vars)
		{
		  if (val_end < val + sizeof val - 1)
		    *val_end++ = ch;
		  UPDATE_BEG_END_STATE (ch);
		  ch = READCHAR;
		}
	      if (! in_file_vars)
		/* The value was terminated by an end-marker, which
		   remove.  */
		val_end -= 3;
	      while (val_end > val
		     && (val_end[-1] == ' ' || val_end[-1] == '\t'))
		val_end--;
	      *val_end = '\0';

	      if (strcmp (var, "lexical-binding") == 0)
		/* This is it...  */
		{
		  rv = (strcmp (val, "nil") != 0);
		  break;
		}
	    }
	}

      while (ch != '\n' && ch != EOF)
	ch = READCHAR;

      return rv;
    }
}

955

Kenichi Handa's avatar
Kenichi Handa committed
956
/* Value is a version number of byte compiled code if the file
Miles Bader's avatar
Miles Bader committed
957
   associated with file descriptor FD is a compiled Lisp file that's
Kenichi Handa's avatar
Kenichi Handa committed
958 959 960
   safe to load.  Only files compiled with Emacs are safe to load.
   Files compiled with XEmacs can lead to a crash in Fbyte_code
   because of an incompatible change in the byte compiler.  */
961 962

static int
963
safe_to_load_p (int fd)
964 965 966 967
{
  char buf[512];
  int nbytes, i;
  int safe_p = 1;
Kenichi Handa's avatar
Kenichi Handa committed
968
  int version = 1;
969 970 971 972 973 974 975 976 977

  /* Read the first few bytes from the file, and look for a line
     specifying the byte compiler version used.  */
  nbytes = emacs_read (fd, buf, sizeof buf - 1);
  if (nbytes > 0)
    {
      buf[nbytes] = '\0';

      /* Skip to the next newline, skipping over the initial `ELC'