| 1 | /* Copyright (C) 1995, 1996, 1997, 1999, 2000, 2001, 2003, 2004, 2006, |
| 2 | * 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation, Inc. |
| 3 | * |
| 4 | * This library is free software; you can redistribute it and/or |
| 5 | * modify it under the terms of the GNU Lesser General Public License |
| 6 | * as published by the Free Software Foundation; either version 3 of |
| 7 | * the License, or (at your option) any later version. |
| 8 | * |
| 9 | * This library is distributed in the hope that it will be useful, but |
| 10 | * WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 12 | * Lesser General Public License for more details. |
| 13 | * |
| 14 | * You should have received a copy of the GNU Lesser General Public |
| 15 | * License along with this library; if not, write to the Free Software |
| 16 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA |
| 17 | * 02110-1301 USA |
| 18 | */ |
| 19 | |
| 20 | |
| 21 | \f |
| 22 | |
| 23 | #ifdef HAVE_CONFIG_H |
| 24 | # include <config.h> |
| 25 | #endif |
| 26 | |
| 27 | #include <stdio.h> |
| 28 | #include <ctype.h> |
| 29 | #include <string.h> |
| 30 | #include <unistd.h> |
| 31 | #include <unicase.h> |
| 32 | #include <unictype.h> |
| 33 | |
| 34 | #include "libguile/_scm.h" |
| 35 | #include "libguile/bytevectors.h" |
| 36 | #include "libguile/chars.h" |
| 37 | #include "libguile/eval.h" |
| 38 | #include "libguile/arrays.h" |
| 39 | #include "libguile/bitvectors.h" |
| 40 | #include "libguile/keywords.h" |
| 41 | #include "libguile/alist.h" |
| 42 | #include "libguile/srcprop.h" |
| 43 | #include "libguile/hashtab.h" |
| 44 | #include "libguile/hash.h" |
| 45 | #include "libguile/ports.h" |
| 46 | #include "libguile/fports.h" |
| 47 | #include "libguile/root.h" |
| 48 | #include "libguile/strings.h" |
| 49 | #include "libguile/strports.h" |
| 50 | #include "libguile/vectors.h" |
| 51 | #include "libguile/validate.h" |
| 52 | #include "libguile/srfi-4.h" |
| 53 | #include "libguile/srfi-13.h" |
| 54 | |
| 55 | #include "libguile/read.h" |
| 56 | #include "libguile/private-options.h" |
| 57 | |
| 58 | |
| 59 | \f |
| 60 | |
| 61 | SCM_GLOBAL_SYMBOL (scm_sym_dot, "."); |
| 62 | SCM_SYMBOL (scm_keyword_prefix, "prefix"); |
| 63 | SCM_SYMBOL (scm_keyword_postfix, "postfix"); |
| 64 | SCM_SYMBOL (sym_nil, "nil"); |
| 65 | |
| 66 | scm_t_option scm_read_opts[] = { |
| 67 | { SCM_OPTION_BOOLEAN, "copy", 0, |
| 68 | "Copy source code expressions." }, |
| 69 | { SCM_OPTION_BOOLEAN, "positions", 1, |
| 70 | "Record positions of source code expressions." }, |
| 71 | { SCM_OPTION_BOOLEAN, "case-insensitive", 0, |
| 72 | "Convert symbols to lower case."}, |
| 73 | { SCM_OPTION_SCM, "keywords", (scm_t_bits) SCM_BOOL_F_BITS, |
| 74 | "Style of keyword recognition: #f, 'prefix or 'postfix."}, |
| 75 | { SCM_OPTION_BOOLEAN, "r6rs-hex-escapes", 0, |
| 76 | "Use R6RS variable-length character and string hex escapes."}, |
| 77 | { SCM_OPTION_BOOLEAN, "square-brackets", 1, |
| 78 | "Treat `[' and `]' as parentheses, for R6RS compatibility."}, |
| 79 | { SCM_OPTION_BOOLEAN, "hungry-eol-escapes", 0, |
| 80 | "In strings, consume leading whitespace after an escaped end-of-line."}, |
| 81 | { 0, }, |
| 82 | }; |
| 83 | |
| 84 | /* |
| 85 | Give meaningful error messages for errors |
| 86 | |
| 87 | We use the format |
| 88 | |
| 89 | FILE:LINE:COL: MESSAGE |
| 90 | This happened in .... |
| 91 | |
| 92 | This is not standard GNU format, but the test-suite likes the real |
| 93 | message to be in front. |
| 94 | |
| 95 | */ |
| 96 | |
| 97 | |
| 98 | void |
| 99 | scm_i_input_error (char const *function, |
| 100 | SCM port, const char *message, SCM arg) |
| 101 | { |
| 102 | SCM fn = (scm_is_string (SCM_FILENAME(port)) |
| 103 | ? SCM_FILENAME(port) |
| 104 | : scm_from_locale_string ("#<unknown port>")); |
| 105 | |
| 106 | SCM string_port = scm_open_output_string (); |
| 107 | SCM string = SCM_EOL; |
| 108 | scm_simple_format (string_port, |
| 109 | scm_from_locale_string ("~A:~S:~S: ~A"), |
| 110 | scm_list_4 (fn, |
| 111 | scm_from_long (SCM_LINUM (port) + 1), |
| 112 | scm_from_int (SCM_COL (port) + 1), |
| 113 | scm_from_locale_string (message))); |
| 114 | |
| 115 | string = scm_get_output_string (string_port); |
| 116 | scm_close_output_port (string_port); |
| 117 | scm_error_scm (scm_from_latin1_symbol ("read-error"), |
| 118 | function? scm_from_locale_string (function) : SCM_BOOL_F, |
| 119 | string, |
| 120 | arg, |
| 121 | SCM_BOOL_F); |
| 122 | } |
| 123 | |
| 124 | |
| 125 | SCM_DEFINE (scm_read_options, "read-options-interface", 0, 1, 0, |
| 126 | (SCM setting), |
| 127 | "Option interface for the read options. Instead of using\n" |
| 128 | "this procedure directly, use the procedures @code{read-enable},\n" |
| 129 | "@code{read-disable}, @code{read-set!} and @code{read-options}.") |
| 130 | #define FUNC_NAME s_scm_read_options |
| 131 | { |
| 132 | SCM ans = scm_options (setting, |
| 133 | scm_read_opts, |
| 134 | FUNC_NAME); |
| 135 | if (SCM_COPY_SOURCE_P) |
| 136 | SCM_RECORD_POSITIONS_P = 1; |
| 137 | return ans; |
| 138 | } |
| 139 | #undef FUNC_NAME |
| 140 | |
| 141 | /* A fluid referring to an association list mapping extra hash |
| 142 | characters to procedures. */ |
| 143 | static SCM *scm_i_read_hash_procedures; |
| 144 | |
| 145 | static SCM |
| 146 | scm_i_read_hash_procedures_ref (void) |
| 147 | { |
| 148 | return scm_fluid_ref (*scm_i_read_hash_procedures); |
| 149 | } |
| 150 | |
| 151 | static void |
| 152 | scm_i_read_hash_procedures_set_x (SCM value) |
| 153 | { |
| 154 | scm_fluid_set_x (*scm_i_read_hash_procedures, value); |
| 155 | } |
| 156 | |
| 157 | \f |
| 158 | /* Token readers. */ |
| 159 | |
| 160 | |
| 161 | /* Size of the C buffer used to read symbols and numbers. */ |
| 162 | #define READER_BUFFER_SIZE 128 |
| 163 | |
| 164 | /* Number of 32-bit codepoints in the buffer used to read strings. */ |
| 165 | #define READER_STRING_BUFFER_SIZE 128 |
| 166 | |
| 167 | /* The maximum size of Scheme character names. */ |
| 168 | #define READER_CHAR_NAME_MAX_SIZE 50 |
| 169 | |
| 170 | |
| 171 | /* `isblank' is only in C99. */ |
| 172 | #define CHAR_IS_BLANK_(_chr) \ |
| 173 | (((_chr) == ' ') || ((_chr) == '\t') || ((_chr) == '\n') \ |
| 174 | || ((_chr) == '\f') || ((_chr) == '\r')) |
| 175 | |
| 176 | #ifdef MSDOS |
| 177 | # define CHAR_IS_BLANK(_chr) \ |
| 178 | ((CHAR_IS_BLANK_ (chr)) || ((_chr) == 26)) |
| 179 | #else |
| 180 | # define CHAR_IS_BLANK CHAR_IS_BLANK_ |
| 181 | #endif |
| 182 | |
| 183 | |
| 184 | /* R5RS one-character delimiters (see section 7.1.1, ``Lexical |
| 185 | structure''). */ |
| 186 | #define CHAR_IS_R5RS_DELIMITER(c) \ |
| 187 | (CHAR_IS_BLANK (c) \ |
| 188 | || (c == ')') || (c == '(') || (c == ';') || (c == '"') \ |
| 189 | || (SCM_SQUARE_BRACKETS_P && ((c == '[') || (c == ']')))) |
| 190 | |
| 191 | #define CHAR_IS_DELIMITER CHAR_IS_R5RS_DELIMITER |
| 192 | |
| 193 | /* Exponent markers, as defined in section 7.1.1 of R5RS, ``Lexical |
| 194 | Structure''. */ |
| 195 | #define CHAR_IS_EXPONENT_MARKER(_chr) \ |
| 196 | (((_chr) == 'e') || ((_chr) == 's') || ((_chr) == 'f') \ |
| 197 | || ((_chr) == 'd') || ((_chr) == 'l')) |
| 198 | |
| 199 | /* Read an SCSH block comment. */ |
| 200 | static SCM scm_read_scsh_block_comment (scm_t_wchar, SCM); |
| 201 | static SCM scm_read_r6rs_block_comment (scm_t_wchar, SCM); |
| 202 | static SCM scm_read_commented_expression (scm_t_wchar, SCM); |
| 203 | static SCM scm_read_shebang (scm_t_wchar, SCM); |
| 204 | static SCM scm_get_hash_procedure (int); |
| 205 | |
| 206 | /* Read from PORT until a delimiter (e.g., a whitespace) is read. Put the |
| 207 | result in the pre-allocated buffer BUF. Return zero if the whole token has |
| 208 | fewer than BUF_SIZE bytes, non-zero otherwise. READ will be set the number of |
| 209 | bytes actually read. */ |
| 210 | static int |
| 211 | read_token (SCM port, char *buf, size_t buf_size, size_t *read) |
| 212 | { |
| 213 | *read = 0; |
| 214 | |
| 215 | while (*read < buf_size) |
| 216 | { |
| 217 | int chr; |
| 218 | |
| 219 | chr = scm_get_byte_or_eof (port); |
| 220 | |
| 221 | if (chr == EOF) |
| 222 | return 0; |
| 223 | else if (CHAR_IS_DELIMITER (chr)) |
| 224 | { |
| 225 | scm_unget_byte (chr, port); |
| 226 | return 0; |
| 227 | } |
| 228 | else |
| 229 | { |
| 230 | *buf = (char) chr; |
| 231 | buf++, (*read)++; |
| 232 | } |
| 233 | } |
| 234 | |
| 235 | return 1; |
| 236 | } |
| 237 | |
| 238 | /* Like `read_token', but return either BUFFER, or a GC-allocated buffer |
| 239 | if the token doesn't fit in BUFFER_SIZE bytes. */ |
| 240 | static char * |
| 241 | read_complete_token (SCM port, char *buffer, size_t buffer_size, |
| 242 | size_t *read) |
| 243 | { |
| 244 | int overflow = 0; |
| 245 | size_t bytes_read, overflow_size = 0; |
| 246 | char *overflow_buffer = NULL; |
| 247 | |
| 248 | do |
| 249 | { |
| 250 | overflow = read_token (port, buffer, buffer_size, &bytes_read); |
| 251 | if (bytes_read == 0) |
| 252 | break; |
| 253 | if (overflow || overflow_size != 0) |
| 254 | { |
| 255 | if (overflow_size == 0) |
| 256 | { |
| 257 | overflow_buffer = scm_gc_malloc_pointerless (bytes_read, "read"); |
| 258 | memcpy (overflow_buffer, buffer, bytes_read); |
| 259 | overflow_size = bytes_read; |
| 260 | } |
| 261 | else |
| 262 | { |
| 263 | char *new_buf = |
| 264 | scm_gc_malloc_pointerless (overflow_size + bytes_read, "read"); |
| 265 | |
| 266 | memcpy (new_buf, overflow_buffer, overflow_size); |
| 267 | memcpy (new_buf + overflow_size, buffer, bytes_read); |
| 268 | |
| 269 | overflow_buffer = new_buf; |
| 270 | overflow_size += bytes_read; |
| 271 | } |
| 272 | } |
| 273 | } |
| 274 | while (overflow); |
| 275 | |
| 276 | if (overflow_size) |
| 277 | *read = overflow_size; |
| 278 | else |
| 279 | *read = bytes_read; |
| 280 | |
| 281 | return (overflow_size > 0 ? overflow_buffer : buffer); |
| 282 | } |
| 283 | |
| 284 | /* Skip whitespace from PORT and return the first non-whitespace character |
| 285 | read. Raise an error on end-of-file. */ |
| 286 | static int |
| 287 | flush_ws (SCM port, const char *eoferr) |
| 288 | { |
| 289 | scm_t_wchar c; |
| 290 | while (1) |
| 291 | switch (c = scm_getc (port)) |
| 292 | { |
| 293 | case EOF: |
| 294 | goteof: |
| 295 | if (eoferr) |
| 296 | { |
| 297 | scm_i_input_error (eoferr, |
| 298 | port, |
| 299 | "end of file", |
| 300 | SCM_EOL); |
| 301 | } |
| 302 | return c; |
| 303 | |
| 304 | case ';': |
| 305 | lp: |
| 306 | switch (c = scm_getc (port)) |
| 307 | { |
| 308 | case EOF: |
| 309 | goto goteof; |
| 310 | default: |
| 311 | goto lp; |
| 312 | case SCM_LINE_INCREMENTORS: |
| 313 | break; |
| 314 | } |
| 315 | break; |
| 316 | |
| 317 | case '#': |
| 318 | switch (c = scm_getc (port)) |
| 319 | { |
| 320 | case EOF: |
| 321 | eoferr = "read_sharp"; |
| 322 | goto goteof; |
| 323 | case '!': |
| 324 | scm_read_shebang (c, port); |
| 325 | break; |
| 326 | case ';': |
| 327 | scm_read_commented_expression (c, port); |
| 328 | break; |
| 329 | case '|': |
| 330 | if (scm_is_false (scm_get_hash_procedure (c))) |
| 331 | { |
| 332 | scm_read_r6rs_block_comment (c, port); |
| 333 | break; |
| 334 | } |
| 335 | /* fall through */ |
| 336 | default: |
| 337 | scm_ungetc (c, port); |
| 338 | return '#'; |
| 339 | } |
| 340 | break; |
| 341 | |
| 342 | case SCM_LINE_INCREMENTORS: |
| 343 | case SCM_SINGLE_SPACES: |
| 344 | case '\t': |
| 345 | break; |
| 346 | |
| 347 | default: |
| 348 | return c; |
| 349 | } |
| 350 | |
| 351 | return 0; |
| 352 | } |
| 353 | |
| 354 | |
| 355 | \f |
| 356 | /* Token readers. */ |
| 357 | |
| 358 | static SCM scm_read_expression (SCM port); |
| 359 | static SCM scm_read_sharp (int chr, SCM port, long line, int column); |
| 360 | |
| 361 | |
| 362 | static SCM |
| 363 | maybe_annotate_source (SCM x, SCM port, long line, int column) |
| 364 | { |
| 365 | if (SCM_RECORD_POSITIONS_P) |
| 366 | scm_i_set_source_properties_x (x, line, column, SCM_FILENAME (port)); |
| 367 | return x; |
| 368 | } |
| 369 | |
| 370 | static SCM |
| 371 | scm_read_sexp (scm_t_wchar chr, SCM port) |
| 372 | #define FUNC_NAME "scm_i_lreadparen" |
| 373 | { |
| 374 | int c; |
| 375 | SCM tmp, tl, ans = SCM_EOL; |
| 376 | const int terminating_char = ((chr == '[') ? ']' : ')'); |
| 377 | |
| 378 | /* Need to capture line and column numbers here. */ |
| 379 | long line = SCM_LINUM (port); |
| 380 | int column = SCM_COL (port) - 1; |
| 381 | |
| 382 | c = flush_ws (port, FUNC_NAME); |
| 383 | if (terminating_char == c) |
| 384 | return SCM_EOL; |
| 385 | |
| 386 | scm_ungetc (c, port); |
| 387 | tmp = scm_read_expression (port); |
| 388 | |
| 389 | /* Note that it is possible for scm_read_expression to return |
| 390 | scm_sym_dot, but not as part of a dotted pair: as in #{.}#. So |
| 391 | check that it's a real dot by checking `c'. */ |
| 392 | if (c == '.' && scm_is_eq (scm_sym_dot, tmp)) |
| 393 | { |
| 394 | ans = scm_read_expression (port); |
| 395 | if (terminating_char != (c = flush_ws (port, FUNC_NAME))) |
| 396 | scm_i_input_error (FUNC_NAME, port, "missing close paren", |
| 397 | SCM_EOL); |
| 398 | return ans; |
| 399 | } |
| 400 | |
| 401 | /* Build the head of the list structure. */ |
| 402 | ans = tl = scm_cons (tmp, SCM_EOL); |
| 403 | |
| 404 | while (terminating_char != (c = flush_ws (port, FUNC_NAME))) |
| 405 | { |
| 406 | SCM new_tail; |
| 407 | |
| 408 | if (c == ')' || (SCM_SQUARE_BRACKETS_P && c == ']')) |
| 409 | scm_i_input_error (FUNC_NAME, port, |
| 410 | "in pair: mismatched close paren: ~A", |
| 411 | scm_list_1 (SCM_MAKE_CHAR (c))); |
| 412 | |
| 413 | scm_ungetc (c, port); |
| 414 | tmp = scm_read_expression (port); |
| 415 | |
| 416 | /* See above note about scm_sym_dot. */ |
| 417 | if (c == '.' && scm_is_eq (scm_sym_dot, tmp)) |
| 418 | { |
| 419 | SCM_SETCDR (tl, scm_read_expression (port)); |
| 420 | |
| 421 | c = flush_ws (port, FUNC_NAME); |
| 422 | if (terminating_char != c) |
| 423 | scm_i_input_error (FUNC_NAME, port, |
| 424 | "in pair: missing close paren", SCM_EOL); |
| 425 | goto exit; |
| 426 | } |
| 427 | |
| 428 | new_tail = scm_cons (tmp, SCM_EOL); |
| 429 | SCM_SETCDR (tl, new_tail); |
| 430 | tl = new_tail; |
| 431 | } |
| 432 | |
| 433 | exit: |
| 434 | return maybe_annotate_source (ans, port, line, column); |
| 435 | } |
| 436 | #undef FUNC_NAME |
| 437 | |
| 438 | |
| 439 | /* Read a hexadecimal number NDIGITS in length. Put its value into the variable |
| 440 | C. If TERMINATOR is non-null, terminate early if the TERMINATOR character is |
| 441 | found. */ |
| 442 | #define SCM_READ_HEX_ESCAPE(ndigits, terminator) \ |
| 443 | do \ |
| 444 | { \ |
| 445 | scm_t_wchar a; \ |
| 446 | size_t i = 0; \ |
| 447 | c = 0; \ |
| 448 | while (i < ndigits) \ |
| 449 | { \ |
| 450 | a = scm_getc (port); \ |
| 451 | if (a == EOF) \ |
| 452 | goto str_eof; \ |
| 453 | if (terminator \ |
| 454 | && (a == (scm_t_wchar) terminator) \ |
| 455 | && (i > 0)) \ |
| 456 | break; \ |
| 457 | if ('0' <= a && a <= '9') \ |
| 458 | a -= '0'; \ |
| 459 | else if ('A' <= a && a <= 'F') \ |
| 460 | a = a - 'A' + 10; \ |
| 461 | else if ('a' <= a && a <= 'f') \ |
| 462 | a = a - 'a' + 10; \ |
| 463 | else \ |
| 464 | { \ |
| 465 | c = a; \ |
| 466 | goto bad_escaped; \ |
| 467 | } \ |
| 468 | c = c * 16 + a; \ |
| 469 | i ++; \ |
| 470 | } \ |
| 471 | } while (0) |
| 472 | |
| 473 | static void |
| 474 | skip_intraline_whitespace (SCM port) |
| 475 | { |
| 476 | scm_t_wchar c; |
| 477 | |
| 478 | do |
| 479 | { |
| 480 | c = scm_getc (port); |
| 481 | if (c == EOF) |
| 482 | return; |
| 483 | } |
| 484 | while (c == '\t' || uc_is_general_category (c, UC_SPACE_SEPARATOR)); |
| 485 | |
| 486 | scm_ungetc (c, port); |
| 487 | } |
| 488 | |
| 489 | static SCM |
| 490 | scm_read_string (int chr, SCM port) |
| 491 | #define FUNC_NAME "scm_lreadr" |
| 492 | { |
| 493 | /* For strings smaller than C_STR, this function creates only one Scheme |
| 494 | object (the string returned). */ |
| 495 | |
| 496 | SCM str = SCM_EOL; |
| 497 | size_t c_str_len = 0; |
| 498 | scm_t_wchar c, c_str[READER_STRING_BUFFER_SIZE]; |
| 499 | |
| 500 | /* Need to capture line and column numbers here. */ |
| 501 | long line = SCM_LINUM (port); |
| 502 | int column = SCM_COL (port) - 1; |
| 503 | |
| 504 | while ('"' != (c = scm_getc (port))) |
| 505 | { |
| 506 | if (c == EOF) |
| 507 | { |
| 508 | str_eof: |
| 509 | scm_i_input_error (FUNC_NAME, port, |
| 510 | "end of file in string constant", SCM_EOL); |
| 511 | } |
| 512 | |
| 513 | if (c_str_len + 1 >= READER_STRING_BUFFER_SIZE) |
| 514 | { |
| 515 | str = scm_cons (scm_from_utf32_stringn (c_str, c_str_len), str); |
| 516 | c_str_len = 0; |
| 517 | } |
| 518 | |
| 519 | if (c == '\\') |
| 520 | { |
| 521 | switch (c = scm_getc (port)) |
| 522 | { |
| 523 | case EOF: |
| 524 | goto str_eof; |
| 525 | case '"': |
| 526 | case '\\': |
| 527 | break; |
| 528 | case '\n': |
| 529 | if (SCM_HUNGRY_EOL_ESCAPES_P) |
| 530 | skip_intraline_whitespace (port); |
| 531 | continue; |
| 532 | case '0': |
| 533 | c = '\0'; |
| 534 | break; |
| 535 | case 'f': |
| 536 | c = '\f'; |
| 537 | break; |
| 538 | case 'n': |
| 539 | c = '\n'; |
| 540 | break; |
| 541 | case 'r': |
| 542 | c = '\r'; |
| 543 | break; |
| 544 | case 't': |
| 545 | c = '\t'; |
| 546 | break; |
| 547 | case 'a': |
| 548 | c = '\007'; |
| 549 | break; |
| 550 | case 'v': |
| 551 | c = '\v'; |
| 552 | break; |
| 553 | case 'b': |
| 554 | c = '\010'; |
| 555 | break; |
| 556 | case 'x': |
| 557 | if (SCM_R6RS_ESCAPES_P) |
| 558 | SCM_READ_HEX_ESCAPE (10, ';'); |
| 559 | else |
| 560 | SCM_READ_HEX_ESCAPE (2, '\0'); |
| 561 | break; |
| 562 | case 'u': |
| 563 | if (!SCM_R6RS_ESCAPES_P) |
| 564 | { |
| 565 | SCM_READ_HEX_ESCAPE (4, '\0'); |
| 566 | break; |
| 567 | } |
| 568 | case 'U': |
| 569 | if (!SCM_R6RS_ESCAPES_P) |
| 570 | { |
| 571 | SCM_READ_HEX_ESCAPE (6, '\0'); |
| 572 | break; |
| 573 | } |
| 574 | default: |
| 575 | bad_escaped: |
| 576 | scm_i_input_error (FUNC_NAME, port, |
| 577 | "illegal character in escape sequence: ~S", |
| 578 | scm_list_1 (SCM_MAKE_CHAR (c))); |
| 579 | } |
| 580 | } |
| 581 | |
| 582 | c_str[c_str_len++] = c; |
| 583 | } |
| 584 | |
| 585 | if (scm_is_null (str)) |
| 586 | /* Fast path: we got a string that fits in C_STR. */ |
| 587 | str = scm_from_utf32_stringn (c_str, c_str_len); |
| 588 | else |
| 589 | { |
| 590 | if (c_str_len > 0) |
| 591 | str = scm_cons (scm_from_utf32_stringn (c_str, c_str_len), str); |
| 592 | |
| 593 | str = scm_string_concatenate_reverse (str, SCM_UNDEFINED, SCM_UNDEFINED); |
| 594 | } |
| 595 | |
| 596 | return maybe_annotate_source (str, port, line, column); |
| 597 | } |
| 598 | #undef FUNC_NAME |
| 599 | |
| 600 | |
| 601 | static SCM |
| 602 | scm_read_number (scm_t_wchar chr, SCM port) |
| 603 | { |
| 604 | SCM result, str = SCM_EOL; |
| 605 | char local_buffer[READER_BUFFER_SIZE], *buffer; |
| 606 | size_t bytes_read; |
| 607 | scm_t_port *pt = SCM_PTAB_ENTRY (port); |
| 608 | |
| 609 | /* Need to capture line and column numbers here. */ |
| 610 | long line = SCM_LINUM (port); |
| 611 | int column = SCM_COL (port) - 1; |
| 612 | |
| 613 | scm_ungetc (chr, port); |
| 614 | buffer = read_complete_token (port, local_buffer, sizeof local_buffer, |
| 615 | &bytes_read); |
| 616 | |
| 617 | str = scm_from_stringn (buffer, bytes_read, pt->encoding, pt->ilseq_handler); |
| 618 | |
| 619 | result = scm_string_to_number (str, SCM_UNDEFINED); |
| 620 | if (scm_is_false (result)) |
| 621 | { |
| 622 | /* Return a symbol instead of a number */ |
| 623 | if (SCM_CASE_INSENSITIVE_P) |
| 624 | str = scm_string_downcase_x (str); |
| 625 | result = scm_string_to_symbol (str); |
| 626 | } |
| 627 | else if (SCM_NIMP (result)) |
| 628 | result = maybe_annotate_source (result, port, line, column); |
| 629 | |
| 630 | SCM_COL (port) += scm_i_string_length (str); |
| 631 | return result; |
| 632 | } |
| 633 | |
| 634 | static SCM |
| 635 | scm_read_mixed_case_symbol (scm_t_wchar chr, SCM port) |
| 636 | { |
| 637 | SCM result; |
| 638 | int ends_with_colon = 0; |
| 639 | size_t bytes_read; |
| 640 | int postfix = scm_is_eq (SCM_PACK (SCM_KEYWORD_STYLE), scm_keyword_postfix); |
| 641 | char local_buffer[READER_BUFFER_SIZE], *buffer; |
| 642 | scm_t_port *pt = SCM_PTAB_ENTRY (port); |
| 643 | SCM str; |
| 644 | |
| 645 | scm_ungetc (chr, port); |
| 646 | buffer = read_complete_token (port, local_buffer, sizeof local_buffer, |
| 647 | &bytes_read); |
| 648 | if (bytes_read > 0) |
| 649 | ends_with_colon = buffer[bytes_read - 1] == ':'; |
| 650 | |
| 651 | if (postfix && ends_with_colon && (bytes_read > 1)) |
| 652 | { |
| 653 | str = scm_from_stringn (buffer, bytes_read - 1, |
| 654 | pt->encoding, pt->ilseq_handler); |
| 655 | |
| 656 | if (SCM_CASE_INSENSITIVE_P) |
| 657 | str = scm_string_downcase_x (str); |
| 658 | result = scm_symbol_to_keyword (scm_string_to_symbol (str)); |
| 659 | } |
| 660 | else |
| 661 | { |
| 662 | str = scm_from_stringn (buffer, bytes_read, |
| 663 | pt->encoding, pt->ilseq_handler); |
| 664 | |
| 665 | if (SCM_CASE_INSENSITIVE_P) |
| 666 | str = scm_string_downcase_x (str); |
| 667 | result = scm_string_to_symbol (str); |
| 668 | } |
| 669 | |
| 670 | SCM_COL (port) += scm_i_string_length (str); |
| 671 | return result; |
| 672 | } |
| 673 | |
| 674 | static SCM |
| 675 | scm_read_number_and_radix (scm_t_wchar chr, SCM port) |
| 676 | #define FUNC_NAME "scm_lreadr" |
| 677 | { |
| 678 | SCM result; |
| 679 | size_t read; |
| 680 | char local_buffer[READER_BUFFER_SIZE], *buffer; |
| 681 | unsigned int radix; |
| 682 | SCM str; |
| 683 | scm_t_port *pt; |
| 684 | |
| 685 | switch (chr) |
| 686 | { |
| 687 | case 'B': |
| 688 | case 'b': |
| 689 | radix = 2; |
| 690 | break; |
| 691 | |
| 692 | case 'o': |
| 693 | case 'O': |
| 694 | radix = 8; |
| 695 | break; |
| 696 | |
| 697 | case 'd': |
| 698 | case 'D': |
| 699 | radix = 10; |
| 700 | break; |
| 701 | |
| 702 | case 'x': |
| 703 | case 'X': |
| 704 | radix = 16; |
| 705 | break; |
| 706 | |
| 707 | default: |
| 708 | scm_ungetc (chr, port); |
| 709 | scm_ungetc ('#', port); |
| 710 | radix = 10; |
| 711 | } |
| 712 | |
| 713 | buffer = read_complete_token (port, local_buffer, sizeof local_buffer, |
| 714 | &read); |
| 715 | |
| 716 | pt = SCM_PTAB_ENTRY (port); |
| 717 | str = scm_from_stringn (buffer, read, pt->encoding, pt->ilseq_handler); |
| 718 | |
| 719 | result = scm_string_to_number (str, scm_from_uint (radix)); |
| 720 | |
| 721 | SCM_COL (port) += scm_i_string_length (str); |
| 722 | |
| 723 | if (scm_is_true (result)) |
| 724 | return result; |
| 725 | |
| 726 | scm_i_input_error (FUNC_NAME, port, "unknown # object", SCM_EOL); |
| 727 | |
| 728 | return SCM_BOOL_F; |
| 729 | } |
| 730 | #undef FUNC_NAME |
| 731 | |
| 732 | static SCM |
| 733 | scm_read_quote (int chr, SCM port) |
| 734 | { |
| 735 | SCM p; |
| 736 | long line = SCM_LINUM (port); |
| 737 | int column = SCM_COL (port) - 1; |
| 738 | |
| 739 | switch (chr) |
| 740 | { |
| 741 | case '`': |
| 742 | p = scm_sym_quasiquote; |
| 743 | break; |
| 744 | |
| 745 | case '\'': |
| 746 | p = scm_sym_quote; |
| 747 | break; |
| 748 | |
| 749 | case ',': |
| 750 | { |
| 751 | scm_t_wchar c; |
| 752 | |
| 753 | c = scm_getc (port); |
| 754 | if ('@' == c) |
| 755 | p = scm_sym_uq_splicing; |
| 756 | else |
| 757 | { |
| 758 | scm_ungetc (c, port); |
| 759 | p = scm_sym_unquote; |
| 760 | } |
| 761 | break; |
| 762 | } |
| 763 | |
| 764 | default: |
| 765 | fprintf (stderr, "%s: unhandled quote character (%i)\n", |
| 766 | "scm_read_quote", chr); |
| 767 | abort (); |
| 768 | } |
| 769 | |
| 770 | p = scm_cons2 (p, scm_read_expression (port), SCM_EOL); |
| 771 | return maybe_annotate_source (p, port, line, column); |
| 772 | } |
| 773 | |
| 774 | SCM_SYMBOL (sym_syntax, "syntax"); |
| 775 | SCM_SYMBOL (sym_quasisyntax, "quasisyntax"); |
| 776 | SCM_SYMBOL (sym_unsyntax, "unsyntax"); |
| 777 | SCM_SYMBOL (sym_unsyntax_splicing, "unsyntax-splicing"); |
| 778 | |
| 779 | static SCM |
| 780 | scm_read_syntax (int chr, SCM port) |
| 781 | { |
| 782 | SCM p; |
| 783 | long line = SCM_LINUM (port); |
| 784 | int column = SCM_COL (port) - 1; |
| 785 | |
| 786 | switch (chr) |
| 787 | { |
| 788 | case '`': |
| 789 | p = sym_quasisyntax; |
| 790 | break; |
| 791 | |
| 792 | case '\'': |
| 793 | p = sym_syntax; |
| 794 | break; |
| 795 | |
| 796 | case ',': |
| 797 | { |
| 798 | int c; |
| 799 | |
| 800 | c = scm_getc (port); |
| 801 | if ('@' == c) |
| 802 | p = sym_unsyntax_splicing; |
| 803 | else |
| 804 | { |
| 805 | scm_ungetc (c, port); |
| 806 | p = sym_unsyntax; |
| 807 | } |
| 808 | break; |
| 809 | } |
| 810 | |
| 811 | default: |
| 812 | fprintf (stderr, "%s: unhandled syntax character (%i)\n", |
| 813 | "scm_read_syntax", chr); |
| 814 | abort (); |
| 815 | } |
| 816 | |
| 817 | p = scm_cons2 (p, scm_read_expression (port), SCM_EOL); |
| 818 | return maybe_annotate_source (p, port, line, column); |
| 819 | } |
| 820 | |
| 821 | static SCM |
| 822 | scm_read_nil (int chr, SCM port) |
| 823 | { |
| 824 | SCM id = scm_read_mixed_case_symbol (chr, port); |
| 825 | |
| 826 | if (!scm_is_eq (id, sym_nil)) |
| 827 | scm_i_input_error ("scm_read_nil", port, |
| 828 | "unexpected input while reading #nil: ~a", |
| 829 | scm_list_1 (id)); |
| 830 | |
| 831 | return SCM_ELISP_NIL; |
| 832 | } |
| 833 | |
| 834 | static SCM |
| 835 | scm_read_semicolon_comment (int chr, SCM port) |
| 836 | { |
| 837 | int c; |
| 838 | |
| 839 | /* We use the get_byte here because there is no need to get the |
| 840 | locale correct with comment input. This presumes that newline |
| 841 | always represents itself no matter what the encoding is. */ |
| 842 | for (c = scm_get_byte_or_eof (port); |
| 843 | (c != EOF) && (c != '\n'); |
| 844 | c = scm_get_byte_or_eof (port)); |
| 845 | |
| 846 | return SCM_UNSPECIFIED; |
| 847 | } |
| 848 | |
| 849 | \f |
| 850 | /* Sharp readers, i.e. readers called after a `#' sign has been read. */ |
| 851 | |
| 852 | static SCM |
| 853 | scm_read_boolean (int chr, SCM port) |
| 854 | { |
| 855 | switch (chr) |
| 856 | { |
| 857 | case 't': |
| 858 | case 'T': |
| 859 | return SCM_BOOL_T; |
| 860 | |
| 861 | case 'f': |
| 862 | case 'F': |
| 863 | return SCM_BOOL_F; |
| 864 | } |
| 865 | |
| 866 | return SCM_UNSPECIFIED; |
| 867 | } |
| 868 | |
| 869 | static SCM |
| 870 | scm_read_character (scm_t_wchar chr, SCM port) |
| 871 | #define FUNC_NAME "scm_lreadr" |
| 872 | { |
| 873 | char buffer[READER_CHAR_NAME_MAX_SIZE]; |
| 874 | SCM charname; |
| 875 | size_t charname_len, bytes_read; |
| 876 | scm_t_wchar cp; |
| 877 | int overflow; |
| 878 | scm_t_port *pt; |
| 879 | |
| 880 | overflow = read_token (port, buffer, READER_CHAR_NAME_MAX_SIZE, &bytes_read); |
| 881 | if (overflow) |
| 882 | scm_i_input_error (FUNC_NAME, port, "character name too long", SCM_EOL); |
| 883 | |
| 884 | if (bytes_read == 0) |
| 885 | { |
| 886 | chr = scm_getc (port); |
| 887 | if (chr == EOF) |
| 888 | scm_i_input_error (FUNC_NAME, port, "unexpected end of file " |
| 889 | "while reading character", SCM_EOL); |
| 890 | |
| 891 | /* CHR must be a token delimiter, like a whitespace. */ |
| 892 | return (SCM_MAKE_CHAR (chr)); |
| 893 | } |
| 894 | |
| 895 | pt = SCM_PTAB_ENTRY (port); |
| 896 | |
| 897 | /* Simple ASCII characters can be processed immediately. Also, simple |
| 898 | ISO-8859-1 characters can be processed immediately if the encoding for this |
| 899 | port is ISO-8859-1. */ |
| 900 | if (bytes_read == 1 && ((unsigned char) buffer[0] <= 127 || pt->encoding == NULL)) |
| 901 | { |
| 902 | SCM_COL (port) += 1; |
| 903 | return SCM_MAKE_CHAR (buffer[0]); |
| 904 | } |
| 905 | |
| 906 | /* Otherwise, convert the buffer into a proper scheme string for |
| 907 | processing. */ |
| 908 | charname = scm_from_stringn (buffer, bytes_read, pt->encoding, |
| 909 | pt->ilseq_handler); |
| 910 | charname_len = scm_i_string_length (charname); |
| 911 | SCM_COL (port) += charname_len; |
| 912 | cp = scm_i_string_ref (charname, 0); |
| 913 | if (charname_len == 1) |
| 914 | return SCM_MAKE_CHAR (cp); |
| 915 | |
| 916 | /* Ignore dotted circles, which may be used to keep combining characters from |
| 917 | combining with the backslash in #\charname. */ |
| 918 | if (cp == SCM_CODEPOINT_DOTTED_CIRCLE && charname_len == 2) |
| 919 | return SCM_MAKE_CHAR (scm_i_string_ref (charname, 1)); |
| 920 | |
| 921 | if (cp >= '0' && cp < '8') |
| 922 | { |
| 923 | /* Dirk:FIXME:: This type of character syntax is not R5RS |
| 924 | * compliant. Further, it should be verified that the constant |
| 925 | * does only consist of octal digits. */ |
| 926 | SCM p = scm_string_to_number (charname, scm_from_uint (8)); |
| 927 | if (SCM_I_INUMP (p)) |
| 928 | { |
| 929 | scm_t_wchar c = scm_to_uint32 (p); |
| 930 | if (SCM_IS_UNICODE_CHAR (c)) |
| 931 | return SCM_MAKE_CHAR (c); |
| 932 | else |
| 933 | scm_i_input_error (FUNC_NAME, port, |
| 934 | "out-of-range octal character escape: ~a", |
| 935 | scm_list_1 (charname)); |
| 936 | } |
| 937 | } |
| 938 | |
| 939 | if (cp == 'x' && (charname_len > 1)) |
| 940 | { |
| 941 | SCM p; |
| 942 | |
| 943 | /* Convert from hex, skipping the initial 'x' character in CHARNAME */ |
| 944 | p = scm_string_to_number (scm_c_substring (charname, 1, charname_len), |
| 945 | scm_from_uint (16)); |
| 946 | if (SCM_I_INUMP (p)) |
| 947 | { |
| 948 | scm_t_wchar c = scm_to_uint32 (p); |
| 949 | if (SCM_IS_UNICODE_CHAR (c)) |
| 950 | return SCM_MAKE_CHAR (c); |
| 951 | else |
| 952 | scm_i_input_error (FUNC_NAME, port, |
| 953 | "out-of-range hex character escape: ~a", |
| 954 | scm_list_1 (charname)); |
| 955 | } |
| 956 | } |
| 957 | |
| 958 | /* The names of characters should never have non-Latin1 |
| 959 | characters. */ |
| 960 | if (scm_i_is_narrow_string (charname) |
| 961 | || scm_i_try_narrow_string (charname)) |
| 962 | { SCM ch = scm_i_charname_to_char (scm_i_string_chars (charname), |
| 963 | charname_len); |
| 964 | if (scm_is_true (ch)) |
| 965 | return ch; |
| 966 | } |
| 967 | |
| 968 | scm_i_input_error (FUNC_NAME, port, "unknown character name ~a", |
| 969 | scm_list_1 (charname)); |
| 970 | |
| 971 | return SCM_UNSPECIFIED; |
| 972 | } |
| 973 | #undef FUNC_NAME |
| 974 | |
| 975 | static SCM |
| 976 | scm_read_keyword (int chr, SCM port) |
| 977 | { |
| 978 | SCM symbol; |
| 979 | |
| 980 | /* Read the symbol that comprises the keyword. Doing this instead of |
| 981 | invoking a specific symbol reader function allows `scm_read_keyword ()' |
| 982 | to adapt to the delimiters currently valid of symbols. |
| 983 | |
| 984 | XXX: This implementation allows sloppy syntaxes like `#: key'. */ |
| 985 | symbol = scm_read_expression (port); |
| 986 | if (!scm_is_symbol (symbol)) |
| 987 | scm_i_input_error ("scm_read_keyword", port, |
| 988 | "keyword prefix `~a' not followed by a symbol: ~s", |
| 989 | scm_list_2 (SCM_MAKE_CHAR (chr), symbol)); |
| 990 | |
| 991 | return (scm_symbol_to_keyword (symbol)); |
| 992 | } |
| 993 | |
| 994 | static SCM |
| 995 | scm_read_vector (int chr, SCM port, long line, int column) |
| 996 | { |
| 997 | /* Note: We call `scm_read_sexp ()' rather than READER here in order to |
| 998 | guarantee that it's going to do what we want. After all, this is an |
| 999 | implementation detail of `scm_read_vector ()', not a desirable |
| 1000 | property. */ |
| 1001 | return maybe_annotate_source (scm_vector (scm_read_sexp (chr, port)), |
| 1002 | port, line, column); |
| 1003 | } |
| 1004 | |
| 1005 | static SCM |
| 1006 | scm_read_array (int chr, SCM port, long line, int column) |
| 1007 | { |
| 1008 | SCM result = scm_i_read_array (port, chr); |
| 1009 | if (scm_is_false (result)) |
| 1010 | return result; |
| 1011 | else |
| 1012 | return maybe_annotate_source (result, port, line, column); |
| 1013 | } |
| 1014 | |
| 1015 | static SCM |
| 1016 | scm_read_srfi4_vector (int chr, SCM port, long line, int column) |
| 1017 | { |
| 1018 | return scm_read_array (chr, port, line, column); |
| 1019 | } |
| 1020 | |
| 1021 | static SCM |
| 1022 | scm_read_bytevector (scm_t_wchar chr, SCM port, long line, int column) |
| 1023 | { |
| 1024 | chr = scm_getc (port); |
| 1025 | if (chr != 'u') |
| 1026 | goto syntax; |
| 1027 | |
| 1028 | chr = scm_getc (port); |
| 1029 | if (chr != '8') |
| 1030 | goto syntax; |
| 1031 | |
| 1032 | chr = scm_getc (port); |
| 1033 | if (chr != '(') |
| 1034 | goto syntax; |
| 1035 | |
| 1036 | return maybe_annotate_source |
| 1037 | (scm_u8_list_to_bytevector (scm_read_sexp (chr, port)), |
| 1038 | port, line, column); |
| 1039 | |
| 1040 | syntax: |
| 1041 | scm_i_input_error ("read_bytevector", port, |
| 1042 | "invalid bytevector prefix", |
| 1043 | SCM_MAKE_CHAR (chr)); |
| 1044 | return SCM_UNSPECIFIED; |
| 1045 | } |
| 1046 | |
| 1047 | static SCM |
| 1048 | scm_read_guile_bit_vector (scm_t_wchar chr, SCM port, long line, int column) |
| 1049 | { |
| 1050 | /* Read the `#*10101'-style read syntax for bit vectors in Guile. This is |
| 1051 | terribly inefficient but who cares? */ |
| 1052 | SCM s_bits = SCM_EOL; |
| 1053 | |
| 1054 | for (chr = scm_getc (port); |
| 1055 | (chr != EOF) && ((chr == '0') || (chr == '1')); |
| 1056 | chr = scm_getc (port)) |
| 1057 | { |
| 1058 | s_bits = scm_cons ((chr == '0') ? SCM_BOOL_F : SCM_BOOL_T, s_bits); |
| 1059 | } |
| 1060 | |
| 1061 | if (chr != EOF) |
| 1062 | scm_ungetc (chr, port); |
| 1063 | |
| 1064 | return maybe_annotate_source |
| 1065 | (scm_bitvector (scm_reverse_x (s_bits, SCM_EOL)), |
| 1066 | port, line, column); |
| 1067 | } |
| 1068 | |
| 1069 | static SCM |
| 1070 | scm_read_scsh_block_comment (scm_t_wchar chr, SCM port) |
| 1071 | { |
| 1072 | int bang_seen = 0; |
| 1073 | |
| 1074 | for (;;) |
| 1075 | { |
| 1076 | int c = scm_getc (port); |
| 1077 | |
| 1078 | if (c == EOF) |
| 1079 | scm_i_input_error ("skip_block_comment", port, |
| 1080 | "unterminated `#! ... !#' comment", SCM_EOL); |
| 1081 | |
| 1082 | if (c == '!') |
| 1083 | bang_seen = 1; |
| 1084 | else if (c == '#' && bang_seen) |
| 1085 | break; |
| 1086 | else |
| 1087 | bang_seen = 0; |
| 1088 | } |
| 1089 | |
| 1090 | return SCM_UNSPECIFIED; |
| 1091 | } |
| 1092 | |
| 1093 | static SCM |
| 1094 | scm_read_shebang (scm_t_wchar chr, SCM port) |
| 1095 | { |
| 1096 | int c = 0; |
| 1097 | if ((c = scm_get_byte_or_eof (port)) != 'r') |
| 1098 | { |
| 1099 | scm_ungetc (c, port); |
| 1100 | return scm_read_scsh_block_comment (chr, port); |
| 1101 | } |
| 1102 | if ((c = scm_get_byte_or_eof (port)) != '6') |
| 1103 | { |
| 1104 | scm_ungetc (c, port); |
| 1105 | scm_ungetc ('r', port); |
| 1106 | return scm_read_scsh_block_comment (chr, port); |
| 1107 | } |
| 1108 | if ((c = scm_get_byte_or_eof (port)) != 'r') |
| 1109 | { |
| 1110 | scm_ungetc (c, port); |
| 1111 | scm_ungetc ('6', port); |
| 1112 | scm_ungetc ('r', port); |
| 1113 | return scm_read_scsh_block_comment (chr, port); |
| 1114 | } |
| 1115 | if ((c = scm_get_byte_or_eof (port)) != 's') |
| 1116 | { |
| 1117 | scm_ungetc (c, port); |
| 1118 | scm_ungetc ('r', port); |
| 1119 | scm_ungetc ('6', port); |
| 1120 | scm_ungetc ('r', port); |
| 1121 | return scm_read_scsh_block_comment (chr, port); |
| 1122 | } |
| 1123 | |
| 1124 | return SCM_UNSPECIFIED; |
| 1125 | } |
| 1126 | |
| 1127 | static SCM |
| 1128 | scm_read_r6rs_block_comment (scm_t_wchar chr, SCM port) |
| 1129 | { |
| 1130 | /* Unlike SCSH-style block comments, SRFI-30/R6RS block comments may be |
| 1131 | nested. So care must be taken. */ |
| 1132 | int nesting_level = 1; |
| 1133 | |
| 1134 | int a = scm_getc (port); |
| 1135 | |
| 1136 | if (a == EOF) |
| 1137 | scm_i_input_error ("scm_read_r6rs_block_comment", port, |
| 1138 | "unterminated `#| ... |#' comment", SCM_EOL); |
| 1139 | |
| 1140 | while (nesting_level > 0) |
| 1141 | { |
| 1142 | int b = scm_getc (port); |
| 1143 | |
| 1144 | if (b == EOF) |
| 1145 | scm_i_input_error ("scm_read_r6rs_block_comment", port, |
| 1146 | "unterminated `#| ... |#' comment", SCM_EOL); |
| 1147 | |
| 1148 | if (a == '|' && b == '#') |
| 1149 | { |
| 1150 | nesting_level--; |
| 1151 | b = EOF; |
| 1152 | } |
| 1153 | else if (a == '#' && b == '|') |
| 1154 | { |
| 1155 | nesting_level++; |
| 1156 | b = EOF; |
| 1157 | } |
| 1158 | |
| 1159 | a = b; |
| 1160 | } |
| 1161 | |
| 1162 | return SCM_UNSPECIFIED; |
| 1163 | } |
| 1164 | |
| 1165 | static SCM |
| 1166 | scm_read_commented_expression (scm_t_wchar chr, SCM port) |
| 1167 | { |
| 1168 | scm_t_wchar c; |
| 1169 | |
| 1170 | c = flush_ws (port, (char *) NULL); |
| 1171 | if (EOF == c) |
| 1172 | scm_i_input_error ("read_commented_expression", port, |
| 1173 | "no expression after #; comment", SCM_EOL); |
| 1174 | scm_ungetc (c, port); |
| 1175 | scm_read_expression (port); |
| 1176 | return SCM_UNSPECIFIED; |
| 1177 | } |
| 1178 | |
| 1179 | static SCM |
| 1180 | scm_read_extended_symbol (scm_t_wchar chr, SCM port) |
| 1181 | { |
| 1182 | /* Guile's extended symbol read syntax looks like this: |
| 1183 | |
| 1184 | #{This is all a symbol name}# |
| 1185 | |
| 1186 | So here, CHR is expected to be `{'. */ |
| 1187 | int saw_brace = 0; |
| 1188 | size_t len = 0; |
| 1189 | SCM buf = scm_i_make_string (1024, NULL, 0); |
| 1190 | |
| 1191 | buf = scm_i_string_start_writing (buf); |
| 1192 | |
| 1193 | while ((chr = scm_getc (port)) != EOF) |
| 1194 | { |
| 1195 | if (saw_brace) |
| 1196 | { |
| 1197 | if (chr == '#') |
| 1198 | { |
| 1199 | break; |
| 1200 | } |
| 1201 | else |
| 1202 | { |
| 1203 | saw_brace = 0; |
| 1204 | scm_i_string_set_x (buf, len++, '}'); |
| 1205 | } |
| 1206 | } |
| 1207 | |
| 1208 | if (chr == '}') |
| 1209 | saw_brace = 1; |
| 1210 | else if (chr == '\\') |
| 1211 | { |
| 1212 | /* It used to be that print.c would print extended-read-syntax |
| 1213 | symbols with backslashes before "non-standard" chars, but |
| 1214 | this routine wouldn't do anything with those escapes. |
| 1215 | Bummer. What we've done is to change print.c to output |
| 1216 | R6RS hex escapes for those characters, relying on the fact |
| 1217 | that the extended read syntax would never put a `\' before |
| 1218 | an `x'. For now, we just ignore other instances of |
| 1219 | backslash in the string. */ |
| 1220 | switch ((chr = scm_getc (port))) |
| 1221 | { |
| 1222 | case EOF: |
| 1223 | goto done; |
| 1224 | case 'x': |
| 1225 | { |
| 1226 | scm_t_wchar c; |
| 1227 | |
| 1228 | SCM_READ_HEX_ESCAPE (10, ';'); |
| 1229 | scm_i_string_set_x (buf, len++, c); |
| 1230 | break; |
| 1231 | |
| 1232 | str_eof: |
| 1233 | chr = EOF; |
| 1234 | goto done; |
| 1235 | |
| 1236 | bad_escaped: |
| 1237 | scm_i_string_stop_writing (); |
| 1238 | scm_i_input_error ("scm_read_extended_symbol", port, |
| 1239 | "illegal character in escape sequence: ~S", |
| 1240 | scm_list_1 (SCM_MAKE_CHAR (c))); |
| 1241 | break; |
| 1242 | } |
| 1243 | default: |
| 1244 | scm_i_string_set_x (buf, len++, chr); |
| 1245 | break; |
| 1246 | } |
| 1247 | } |
| 1248 | else |
| 1249 | scm_i_string_set_x (buf, len++, chr); |
| 1250 | |
| 1251 | if (len >= scm_i_string_length (buf) - 2) |
| 1252 | { |
| 1253 | SCM addy; |
| 1254 | |
| 1255 | scm_i_string_stop_writing (); |
| 1256 | addy = scm_i_make_string (1024, NULL, 0); |
| 1257 | buf = scm_string_append (scm_list_2 (buf, addy)); |
| 1258 | len = 0; |
| 1259 | buf = scm_i_string_start_writing (buf); |
| 1260 | } |
| 1261 | } |
| 1262 | |
| 1263 | done: |
| 1264 | scm_i_string_stop_writing (); |
| 1265 | if (chr == EOF) |
| 1266 | scm_i_input_error ("scm_read_extended_symbol", port, |
| 1267 | "end of file while reading symbol", SCM_EOL); |
| 1268 | |
| 1269 | return (scm_string_to_symbol (scm_c_substring (buf, 0, len))); |
| 1270 | } |
| 1271 | |
| 1272 | |
| 1273 | \f |
| 1274 | /* Top-level token readers, i.e., dispatchers. */ |
| 1275 | |
| 1276 | static SCM |
| 1277 | scm_read_sharp_extension (int chr, SCM port) |
| 1278 | { |
| 1279 | SCM proc; |
| 1280 | |
| 1281 | proc = scm_get_hash_procedure (chr); |
| 1282 | if (scm_is_true (scm_procedure_p (proc))) |
| 1283 | { |
| 1284 | long line = SCM_LINUM (port); |
| 1285 | int column = SCM_COL (port) - 2; |
| 1286 | SCM got; |
| 1287 | |
| 1288 | got = scm_call_2 (proc, SCM_MAKE_CHAR (chr), port); |
| 1289 | |
| 1290 | if (scm_is_pair (got) && !scm_i_has_source_properties (got)) |
| 1291 | scm_i_set_source_properties_x (got, line, column, SCM_FILENAME (port)); |
| 1292 | |
| 1293 | return got; |
| 1294 | } |
| 1295 | |
| 1296 | return SCM_UNSPECIFIED; |
| 1297 | } |
| 1298 | |
| 1299 | /* The reader for the sharp `#' character. It basically dispatches reads |
| 1300 | among the above token readers. */ |
| 1301 | static SCM |
| 1302 | scm_read_sharp (scm_t_wchar chr, SCM port, long line, int column) |
| 1303 | #define FUNC_NAME "scm_lreadr" |
| 1304 | { |
| 1305 | SCM result; |
| 1306 | |
| 1307 | chr = scm_getc (port); |
| 1308 | |
| 1309 | result = scm_read_sharp_extension (chr, port); |
| 1310 | if (!scm_is_eq (result, SCM_UNSPECIFIED)) |
| 1311 | return result; |
| 1312 | |
| 1313 | switch (chr) |
| 1314 | { |
| 1315 | case '\\': |
| 1316 | return (scm_read_character (chr, port)); |
| 1317 | case '(': |
| 1318 | return (scm_read_vector (chr, port, line, column)); |
| 1319 | case 's': |
| 1320 | case 'u': |
| 1321 | case 'f': |
| 1322 | case 'c': |
| 1323 | /* This one may return either a boolean or an SRFI-4 vector. */ |
| 1324 | return (scm_read_srfi4_vector (chr, port, line, column)); |
| 1325 | case 'v': |
| 1326 | return (scm_read_bytevector (chr, port, line, column)); |
| 1327 | case '*': |
| 1328 | return (scm_read_guile_bit_vector (chr, port, line, column)); |
| 1329 | case 't': |
| 1330 | case 'T': |
| 1331 | case 'F': |
| 1332 | return (scm_read_boolean (chr, port)); |
| 1333 | case ':': |
| 1334 | return (scm_read_keyword (chr, port)); |
| 1335 | case '0': case '1': case '2': case '3': case '4': |
| 1336 | case '5': case '6': case '7': case '8': case '9': |
| 1337 | case '@': |
| 1338 | #if SCM_ENABLE_DEPRECATED |
| 1339 | /* See below for 'i' and 'e'. */ |
| 1340 | case 'a': |
| 1341 | case 'y': |
| 1342 | case 'h': |
| 1343 | case 'l': |
| 1344 | #endif |
| 1345 | return (scm_read_array (chr, port, line, column)); |
| 1346 | |
| 1347 | case 'i': |
| 1348 | case 'e': |
| 1349 | #if SCM_ENABLE_DEPRECATED |
| 1350 | { |
| 1351 | /* When next char is '(', it really is an old-style |
| 1352 | uniform array. */ |
| 1353 | scm_t_wchar next_c = scm_getc (port); |
| 1354 | if (next_c != EOF) |
| 1355 | scm_ungetc (next_c, port); |
| 1356 | if (next_c == '(') |
| 1357 | return scm_read_array (chr, port, line, column); |
| 1358 | /* Fall through. */ |
| 1359 | } |
| 1360 | #endif |
| 1361 | case 'b': |
| 1362 | case 'B': |
| 1363 | case 'o': |
| 1364 | case 'O': |
| 1365 | case 'd': |
| 1366 | case 'D': |
| 1367 | case 'x': |
| 1368 | case 'X': |
| 1369 | case 'I': |
| 1370 | case 'E': |
| 1371 | return (scm_read_number_and_radix (chr, port)); |
| 1372 | case '{': |
| 1373 | return (scm_read_extended_symbol (chr, port)); |
| 1374 | case '!': |
| 1375 | return (scm_read_shebang (chr, port)); |
| 1376 | case ';': |
| 1377 | return (scm_read_commented_expression (chr, port)); |
| 1378 | case '`': |
| 1379 | case '\'': |
| 1380 | case ',': |
| 1381 | return (scm_read_syntax (chr, port)); |
| 1382 | case 'n': |
| 1383 | return (scm_read_nil (chr, port)); |
| 1384 | default: |
| 1385 | result = scm_read_sharp_extension (chr, port); |
| 1386 | if (scm_is_eq (result, SCM_UNSPECIFIED)) |
| 1387 | { |
| 1388 | /* To remain compatible with 1.8 and earlier, the following |
| 1389 | characters have lower precedence than `read-hash-extend' |
| 1390 | characters. */ |
| 1391 | switch (chr) |
| 1392 | { |
| 1393 | case '|': |
| 1394 | return scm_read_r6rs_block_comment (chr, port); |
| 1395 | default: |
| 1396 | scm_i_input_error (FUNC_NAME, port, "Unknown # object: ~S", |
| 1397 | scm_list_1 (SCM_MAKE_CHAR (chr))); |
| 1398 | } |
| 1399 | } |
| 1400 | else |
| 1401 | return result; |
| 1402 | } |
| 1403 | |
| 1404 | return SCM_UNSPECIFIED; |
| 1405 | } |
| 1406 | #undef FUNC_NAME |
| 1407 | |
| 1408 | static SCM |
| 1409 | scm_read_expression (SCM port) |
| 1410 | #define FUNC_NAME "scm_read_expression" |
| 1411 | { |
| 1412 | while (1) |
| 1413 | { |
| 1414 | scm_t_wchar chr; |
| 1415 | |
| 1416 | chr = scm_getc (port); |
| 1417 | |
| 1418 | switch (chr) |
| 1419 | { |
| 1420 | case SCM_WHITE_SPACES: |
| 1421 | case SCM_LINE_INCREMENTORS: |
| 1422 | break; |
| 1423 | case ';': |
| 1424 | (void) scm_read_semicolon_comment (chr, port); |
| 1425 | break; |
| 1426 | case '[': |
| 1427 | if (!SCM_SQUARE_BRACKETS_P) |
| 1428 | return (scm_read_mixed_case_symbol (chr, port)); |
| 1429 | /* otherwise fall through */ |
| 1430 | case '(': |
| 1431 | return (scm_read_sexp (chr, port)); |
| 1432 | case '"': |
| 1433 | return (scm_read_string (chr, port)); |
| 1434 | case '\'': |
| 1435 | case '`': |
| 1436 | case ',': |
| 1437 | return (scm_read_quote (chr, port)); |
| 1438 | case '#': |
| 1439 | { |
| 1440 | long line = SCM_LINUM (port); |
| 1441 | int column = SCM_COL (port) - 1; |
| 1442 | SCM result = scm_read_sharp (chr, port, line, column); |
| 1443 | if (scm_is_eq (result, SCM_UNSPECIFIED)) |
| 1444 | /* We read a comment or some such. */ |
| 1445 | break; |
| 1446 | else |
| 1447 | return result; |
| 1448 | } |
| 1449 | case ')': |
| 1450 | scm_i_input_error (FUNC_NAME, port, "unexpected \")\"", SCM_EOL); |
| 1451 | break; |
| 1452 | case ']': |
| 1453 | if (SCM_SQUARE_BRACKETS_P) |
| 1454 | scm_i_input_error (FUNC_NAME, port, "unexpected \"]\"", SCM_EOL); |
| 1455 | /* otherwise fall through */ |
| 1456 | case EOF: |
| 1457 | return SCM_EOF_VAL; |
| 1458 | case ':': |
| 1459 | if (scm_is_eq (SCM_PACK (SCM_KEYWORD_STYLE), scm_keyword_prefix)) |
| 1460 | return scm_symbol_to_keyword (scm_read_expression (port)); |
| 1461 | /* Fall through. */ |
| 1462 | |
| 1463 | default: |
| 1464 | { |
| 1465 | if (((chr >= '0') && (chr <= '9')) |
| 1466 | || (strchr ("+-.", chr))) |
| 1467 | return (scm_read_number (chr, port)); |
| 1468 | else |
| 1469 | return (scm_read_mixed_case_symbol (chr, port)); |
| 1470 | } |
| 1471 | } |
| 1472 | } |
| 1473 | } |
| 1474 | #undef FUNC_NAME |
| 1475 | |
| 1476 | \f |
| 1477 | /* Actual reader. */ |
| 1478 | |
| 1479 | SCM_DEFINE (scm_read, "read", 0, 1, 0, |
| 1480 | (SCM port), |
| 1481 | "Read an s-expression from the input port @var{port}, or from\n" |
| 1482 | "the current input port if @var{port} is not specified.\n" |
| 1483 | "Any whitespace before the next token is discarded.") |
| 1484 | #define FUNC_NAME s_scm_read |
| 1485 | { |
| 1486 | int c; |
| 1487 | |
| 1488 | if (SCM_UNBNDP (port)) |
| 1489 | port = scm_current_input_port (); |
| 1490 | SCM_VALIDATE_OPINPORT (1, port); |
| 1491 | |
| 1492 | c = flush_ws (port, (char *) NULL); |
| 1493 | if (EOF == c) |
| 1494 | return SCM_EOF_VAL; |
| 1495 | scm_ungetc (c, port); |
| 1496 | |
| 1497 | return (scm_read_expression (port)); |
| 1498 | } |
| 1499 | #undef FUNC_NAME |
| 1500 | |
| 1501 | |
| 1502 | \f |
| 1503 | |
| 1504 | /* Manipulate the read-hash-procedures alist. This could be written in |
| 1505 | Scheme, but maybe it will also be used by C code during initialisation. */ |
| 1506 | SCM_DEFINE (scm_read_hash_extend, "read-hash-extend", 2, 0, 0, |
| 1507 | (SCM chr, SCM proc), |
| 1508 | "Install the procedure @var{proc} for reading expressions\n" |
| 1509 | "starting with the character sequence @code{#} and @var{chr}.\n" |
| 1510 | "@var{proc} will be called with two arguments: the character\n" |
| 1511 | "@var{chr} and the port to read further data from. The object\n" |
| 1512 | "returned will be the return value of @code{read}. \n" |
| 1513 | "Passing @code{#f} for @var{proc} will remove a previous setting. \n" |
| 1514 | ) |
| 1515 | #define FUNC_NAME s_scm_read_hash_extend |
| 1516 | { |
| 1517 | SCM this; |
| 1518 | SCM prev; |
| 1519 | |
| 1520 | SCM_VALIDATE_CHAR (1, chr); |
| 1521 | SCM_ASSERT (scm_is_false (proc) |
| 1522 | || scm_is_eq (scm_procedure_p (proc), SCM_BOOL_T), |
| 1523 | proc, SCM_ARG2, FUNC_NAME); |
| 1524 | |
| 1525 | /* Check if chr is already in the alist. */ |
| 1526 | this = scm_i_read_hash_procedures_ref (); |
| 1527 | prev = SCM_BOOL_F; |
| 1528 | while (1) |
| 1529 | { |
| 1530 | if (scm_is_null (this)) |
| 1531 | { |
| 1532 | /* not found, so add it to the beginning. */ |
| 1533 | if (scm_is_true (proc)) |
| 1534 | { |
| 1535 | SCM new = scm_cons (scm_cons (chr, proc), |
| 1536 | scm_i_read_hash_procedures_ref ()); |
| 1537 | scm_i_read_hash_procedures_set_x (new); |
| 1538 | } |
| 1539 | break; |
| 1540 | } |
| 1541 | if (scm_is_eq (chr, SCM_CAAR (this))) |
| 1542 | { |
| 1543 | /* already in the alist. */ |
| 1544 | if (scm_is_false (proc)) |
| 1545 | { |
| 1546 | /* remove it. */ |
| 1547 | if (scm_is_false (prev)) |
| 1548 | { |
| 1549 | SCM rest = SCM_CDR (scm_i_read_hash_procedures_ref ()); |
| 1550 | scm_i_read_hash_procedures_set_x (rest); |
| 1551 | } |
| 1552 | else |
| 1553 | scm_set_cdr_x (prev, SCM_CDR (this)); |
| 1554 | } |
| 1555 | else |
| 1556 | { |
| 1557 | /* replace it. */ |
| 1558 | scm_set_cdr_x (SCM_CAR (this), proc); |
| 1559 | } |
| 1560 | break; |
| 1561 | } |
| 1562 | prev = this; |
| 1563 | this = SCM_CDR (this); |
| 1564 | } |
| 1565 | |
| 1566 | return SCM_UNSPECIFIED; |
| 1567 | } |
| 1568 | #undef FUNC_NAME |
| 1569 | |
| 1570 | /* Recover the read-hash procedure corresponding to char c. */ |
| 1571 | static SCM |
| 1572 | scm_get_hash_procedure (int c) |
| 1573 | { |
| 1574 | SCM rest = scm_i_read_hash_procedures_ref (); |
| 1575 | |
| 1576 | while (1) |
| 1577 | { |
| 1578 | if (scm_is_null (rest)) |
| 1579 | return SCM_BOOL_F; |
| 1580 | |
| 1581 | if (SCM_CHAR (SCM_CAAR (rest)) == c) |
| 1582 | return SCM_CDAR (rest); |
| 1583 | |
| 1584 | rest = SCM_CDR (rest); |
| 1585 | } |
| 1586 | } |
| 1587 | |
| 1588 | #define SCM_ENCODING_SEARCH_SIZE (500) |
| 1589 | |
| 1590 | /* Search the first few hundred characters of a file for an Emacs-like coding |
| 1591 | declaration. Returns either NULL or a string whose storage has been |
| 1592 | allocated with `scm_gc_malloc ()'. */ |
| 1593 | char * |
| 1594 | scm_i_scan_for_encoding (SCM port) |
| 1595 | { |
| 1596 | scm_t_port *pt; |
| 1597 | char header[SCM_ENCODING_SEARCH_SIZE+1]; |
| 1598 | size_t bytes_read, encoding_length, i; |
| 1599 | char *encoding = NULL; |
| 1600 | int utf8_bom = 0; |
| 1601 | char *pos, *encoding_start; |
| 1602 | int in_comment; |
| 1603 | |
| 1604 | pt = SCM_PTAB_ENTRY (port); |
| 1605 | |
| 1606 | if (pt->rw_active == SCM_PORT_WRITE) |
| 1607 | scm_flush (port); |
| 1608 | |
| 1609 | if (pt->rw_random) |
| 1610 | pt->rw_active = SCM_PORT_READ; |
| 1611 | |
| 1612 | if (pt->read_pos == pt->read_end) |
| 1613 | { |
| 1614 | /* We can use the read buffer, and thus avoid a seek. */ |
| 1615 | if (scm_fill_input (port) == EOF) |
| 1616 | return NULL; |
| 1617 | |
| 1618 | bytes_read = pt->read_end - pt->read_pos; |
| 1619 | if (bytes_read > SCM_ENCODING_SEARCH_SIZE) |
| 1620 | bytes_read = SCM_ENCODING_SEARCH_SIZE; |
| 1621 | |
| 1622 | if (bytes_read <= 1) |
| 1623 | /* An unbuffered port -- don't scan. */ |
| 1624 | return NULL; |
| 1625 | |
| 1626 | memcpy (header, pt->read_pos, bytes_read); |
| 1627 | header[bytes_read] = '\0'; |
| 1628 | } |
| 1629 | else |
| 1630 | { |
| 1631 | /* Try to read some bytes and then seek back. Not all ports |
| 1632 | support seeking back; and indeed some file ports (like |
| 1633 | /dev/urandom) will succeed on an lseek (fd, 0, SEEK_CUR)---the |
| 1634 | check performed by SCM_FPORT_FDES---but fail to seek |
| 1635 | backwards. Hence this block comes second. We prefer to use |
| 1636 | the read buffer in-place. */ |
| 1637 | if (SCM_FPORTP (port) && !SCM_FDES_RANDOM_P (SCM_FPORT_FDES (port))) |
| 1638 | return NULL; |
| 1639 | |
| 1640 | bytes_read = scm_c_read (port, header, SCM_ENCODING_SEARCH_SIZE); |
| 1641 | header[bytes_read] = '\0'; |
| 1642 | scm_seek (port, scm_from_int (0), scm_from_int (SEEK_SET)); |
| 1643 | } |
| 1644 | |
| 1645 | if (bytes_read > 3 |
| 1646 | && header[0] == '\xef' && header[1] == '\xbb' && header[2] == '\xbf') |
| 1647 | utf8_bom = 1; |
| 1648 | |
| 1649 | /* search past "coding[:=]" */ |
| 1650 | pos = header; |
| 1651 | while (1) |
| 1652 | { |
| 1653 | if ((pos = strstr(pos, "coding")) == NULL) |
| 1654 | return NULL; |
| 1655 | |
| 1656 | pos += strlen("coding"); |
| 1657 | if (pos - header >= SCM_ENCODING_SEARCH_SIZE || |
| 1658 | (*pos == ':' || *pos == '=')) |
| 1659 | { |
| 1660 | pos ++; |
| 1661 | break; |
| 1662 | } |
| 1663 | } |
| 1664 | |
| 1665 | /* skip spaces */ |
| 1666 | while (pos - header <= SCM_ENCODING_SEARCH_SIZE && |
| 1667 | (*pos == ' ' || *pos == '\t')) |
| 1668 | pos ++; |
| 1669 | |
| 1670 | /* grab the next token */ |
| 1671 | encoding_start = pos; |
| 1672 | i = 0; |
| 1673 | while (encoding_start + i - header <= SCM_ENCODING_SEARCH_SIZE |
| 1674 | && encoding_start + i - header < bytes_read |
| 1675 | && (isalnum ((int) encoding_start[i]) |
| 1676 | || strchr ("_-.:/,+=()", encoding_start[i]) != NULL)) |
| 1677 | i++; |
| 1678 | |
| 1679 | encoding_length = i; |
| 1680 | if (encoding_length == 0) |
| 1681 | return NULL; |
| 1682 | |
| 1683 | encoding = scm_gc_strndup (encoding_start, encoding_length, "encoding"); |
| 1684 | for (i = 0; i < encoding_length; i++) |
| 1685 | encoding[i] = toupper ((int) encoding[i]); |
| 1686 | |
| 1687 | /* push backwards to make sure we were in a comment */ |
| 1688 | in_comment = 0; |
| 1689 | pos = encoding_start; |
| 1690 | while (pos >= header) |
| 1691 | { |
| 1692 | if (*pos == ';') |
| 1693 | { |
| 1694 | in_comment = 1; |
| 1695 | break; |
| 1696 | } |
| 1697 | else if (*pos == '\n' || pos == header) |
| 1698 | { |
| 1699 | /* This wasn't in a semicolon comment. Check for a |
| 1700 | hash-bang comment. */ |
| 1701 | char *beg = strstr (header, "#!"); |
| 1702 | char *end = strstr (header, "!#"); |
| 1703 | if (beg < encoding_start && encoding_start + encoding_length <= end) |
| 1704 | in_comment = 1; |
| 1705 | break; |
| 1706 | } |
| 1707 | else |
| 1708 | { |
| 1709 | pos --; |
| 1710 | continue; |
| 1711 | } |
| 1712 | } |
| 1713 | if (!in_comment) |
| 1714 | /* This wasn't in a comment */ |
| 1715 | return NULL; |
| 1716 | |
| 1717 | if (utf8_bom && strcmp(encoding, "UTF-8")) |
| 1718 | scm_misc_error (NULL, |
| 1719 | "the port input declares the encoding ~s but is encoded as UTF-8", |
| 1720 | scm_list_1 (scm_from_locale_string (encoding))); |
| 1721 | |
| 1722 | return encoding; |
| 1723 | } |
| 1724 | |
| 1725 | SCM_DEFINE (scm_file_encoding, "file-encoding", 1, 0, 0, |
| 1726 | (SCM port), |
| 1727 | "Scans the port for an Emacs-like character coding declaration\n" |
| 1728 | "near the top of the contents of a port with random-accessible contents.\n" |
| 1729 | "The coding declaration is of the form\n" |
| 1730 | "@code{coding: XXXXX} and must appear in a scheme comment.\n" |
| 1731 | "\n" |
| 1732 | "Returns a string containing the character encoding of the file\n" |
| 1733 | "if a declaration was found, or @code{#f} otherwise.\n") |
| 1734 | #define FUNC_NAME s_scm_file_encoding |
| 1735 | { |
| 1736 | char *enc; |
| 1737 | SCM s_enc; |
| 1738 | |
| 1739 | SCM_VALIDATE_OPINPORT (SCM_ARG1, port); |
| 1740 | |
| 1741 | enc = scm_i_scan_for_encoding (port); |
| 1742 | if (enc == NULL) |
| 1743 | return SCM_BOOL_F; |
| 1744 | else |
| 1745 | { |
| 1746 | s_enc = scm_from_locale_string (enc); |
| 1747 | return s_enc; |
| 1748 | } |
| 1749 | |
| 1750 | return SCM_BOOL_F; |
| 1751 | } |
| 1752 | #undef FUNC_NAME |
| 1753 | |
| 1754 | void |
| 1755 | scm_init_read () |
| 1756 | { |
| 1757 | SCM read_hash_procs; |
| 1758 | |
| 1759 | read_hash_procs = scm_make_fluid_with_default (SCM_EOL); |
| 1760 | |
| 1761 | scm_i_read_hash_procedures = |
| 1762 | SCM_VARIABLE_LOC (scm_c_define ("%read-hash-procedures", read_hash_procs)); |
| 1763 | |
| 1764 | scm_init_opts (scm_read_options, scm_read_opts); |
| 1765 | #include "libguile/read.x" |
| 1766 | } |
| 1767 | |
| 1768 | /* |
| 1769 | Local Variables: |
| 1770 | c-file-style: "gnu" |
| 1771 | End: |
| 1772 | */ |