1 /* Copyright (C) 1995,1996,1997, 1999 Free Software Foundation, Inc.
3 * This program is free software; you can redistribute it and/or modify
4 * it under the terms of the GNU General Public License as published by
5 * the Free Software Foundation; either version 2, or (at your option)
8 * This program is distributed in the hope that it will be useful,
9 * but WITHOUT ANY WARRANTY; without even the implied warranty of
10 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 * GNU General Public License for more details.
13 * You should have received a copy of the GNU General Public License
14 * along with this software; see the file COPYING. If not, write to
15 * the Free Software Foundation, Inc., 59 Temple Place, Suite 330,
16 * Boston, MA 02111-1307 USA
18 * As a special exception, the Free Software Foundation gives permission
19 * for additional uses of the text contained in its release of GUILE.
21 * The exception is that, if you link the GUILE library with other files
22 * to produce an executable, this does not by itself cause the
23 * resulting executable to be covered by the GNU General Public License.
24 * Your use of that executable is in no way restricted on account of
25 * linking the GUILE library code into it.
27 * This exception does not however invalidate any other reasons why
28 * the executable file might be covered by the GNU General Public License.
30 * This exception applies only to the code released by the
31 * Free Software Foundation under the name GUILE. If you copy
32 * code from other Free Software Foundation releases into a copy of
33 * GUILE, as the General Public License permits, the exception does
34 * not apply to the code that you add in this way. To avoid misleading
35 * anyone as to the status of such modified files, you must delete
36 * this exception notice from them.
38 * If you write modifications of your own for GUILE, it is your choice
39 * whether to permit this exception to apply to your modifications.
40 * If you do not wish that, delete this exception notice. */
59 SCM_SYMBOL (scm_keyword_prefix
, "prefix");
61 scm_option scm_read_opts
[] = {
62 { SCM_OPTION_BOOLEAN
, "copy", 0,
63 "Copy source code expressions." },
64 { SCM_OPTION_BOOLEAN
, "positions", 0,
65 "Record positions of source code expressions." },
66 { SCM_OPTION_BOOLEAN
, "case-insensitive", 0,
67 "Convert symbols to lower case."},
68 { SCM_OPTION_SCM
, "keywords", SCM_BOOL_F
,
69 "Style of keyword recognition: #f or 'prefix"}
72 SCM_PROC (s_read_options
, "read-options-interface", 0, 1, 0, scm_read_options
);
75 scm_read_options (setting
)
78 SCM ans
= scm_options (setting
,
82 if (SCM_COPY_SOURCE_P
)
83 SCM_RECORD_POSITIONS_P
= 1;
87 /* An association list mapping extra hash characters to procedures. */
88 static SCM
*scm_read_hash_procedures
;
90 SCM_PROC (s_read
, "read", 0, 1, 0, scm_read
);
99 if (SCM_UNBNDP (port
))
101 SCM_ASSERT (SCM_NIMP (port
) && SCM_OPINPORTP (port
),
106 c
= scm_flush_ws (port
, (char *) NULL
);
109 scm_ungetc (c
, port
);
111 tok_buf
= scm_makstr (30L, 0);
112 return scm_lreadr (&tok_buf
, port
, ©
);
118 scm_grow_tok_buf (tok_buf
)
121 scm_vector_set_length_x (*tok_buf
, SCM_MAKINUM (2 * SCM_LENGTH (*tok_buf
)));
122 return SCM_CHARS (*tok_buf
);
128 scm_flush_ws (port
, eoferr
)
134 switch (c
= scm_getc (port
))
139 scm_wta (SCM_UNDEFINED
, "end of file in ", eoferr
);
143 switch (c
= scm_getc (port
))
149 case SCM_LINE_INCREMENTORS
:
153 case SCM_LINE_INCREMENTORS
:
154 case SCM_SINGLE_SPACES
:
165 scm_casei_streq (s1
, s2
)
170 if (scm_downcase((int)*s1
) != scm_downcase((int)*s2
))
177 return !(*s1
|| *s2
);
181 /* recsexpr is used when recording expressions
182 * constructed by read:sharp.
184 #ifndef DEBUG_EXTENSIONS
185 #define recsexpr(obj, line, column, filename) (obj)
187 static SCM recsexpr
SCM_P ((SCM obj
, int line
, int column
, SCM filename
));
190 recsexpr (obj
, line
, column
, filename
)
196 if (SCM_IMP (obj
) || SCM_NCONSP(obj
))
200 /* If this sexpr is visible in the read:sharp source, we want to
201 keep that information, so only record non-constant cons cells
202 which haven't previously been read by the reader. */
203 if (SCM_FALSEP (scm_whash_lookup (scm_source_whash
, obj
)))
205 if (SCM_COPY_SOURCE_P
)
207 copy
= scm_cons (recsexpr (SCM_CAR (obj
), line
, column
, filename
),
209 while (SCM_NIMP (tmp
= SCM_CDR (tmp
)) && SCM_CONSP (tmp
))
211 SCM_SETCDR (copy
, scm_cons (recsexpr (SCM_CAR (tmp
),
216 copy
= SCM_CDR (copy
);
218 SCM_SETCDR (copy
, tmp
);
222 recsexpr (SCM_CAR (obj
), line
, column
, filename
);
223 while (SCM_NIMP (tmp
= SCM_CDR (tmp
)) && SCM_CONSP (tmp
))
224 recsexpr (SCM_CAR (tmp
), line
, column
, filename
);
225 copy
= SCM_UNDEFINED
;
227 scm_whash_insert (scm_source_whash
,
229 scm_make_srcprops (line
,
240 /* Consume an SCSH-style block comment. Assume that we've already
241 read the initial `#!', and eat characters until we get a
242 newline/exclamation-point/sharp-sign/newline sequence. */
245 skip_scsh_block_comment (port
)
248 /* Is this portable? Dear God, spare me from the non-eight-bit
249 characters. But is it tasteful? */
254 int c
= scm_getc (port
);
257 scm_wta (SCM_UNDEFINED
,
258 "unterminated `#! ... !#' comment", "read");
259 history
= ((history
<< 8) | (c
& 0xff)) & 0xffffffff;
261 /* Were the last four characters read "\n!#\n"? */
262 if (history
== (('\n' << 24) | ('!' << 16) | ('#' << 8) | '\n'))
268 scm_get_hash_procedure
SCM_P ((int c
));
270 static char s_list
[]="list";
273 scm_lreadr (tok_buf
, port
, copy
)
283 c
= scm_flush_ws (port
, s_read
);
284 tryagain_no_flush_ws
:
291 return SCM_RECORD_POSITIONS_P
292 ? scm_lreadrecparen (tok_buf
, port
, s_list
, copy
)
293 : scm_lreadparen (tok_buf
, port
, s_list
, copy
);
295 scm_wta (SCM_UNDEFINED
, "unexpected \")\"", "read");
302 p
= scm_sym_quasiquote
;
307 p
= scm_sym_uq_splicing
;
310 scm_ungetc (c
, port
);
315 scm_lreadr (tok_buf
, port
, copy
),
317 if (SCM_RECORD_POSITIONS_P
)
318 scm_whash_insert (scm_source_whash
,
320 scm_make_srcprops (SCM_LINUM (port
),
324 ? (*copy
= scm_cons2 (SCM_CAR (p
),
325 SCM_CAR (SCM_CDR (p
)),
335 p
= scm_lreadparen (tok_buf
, port
, "vector", copy
);
336 return SCM_NULLP (p
) ? scm_nullvect
: scm_vector (p
);
357 scm_ungetc (c
, port
);
362 /* start of a shell script. Parse as a block comment,
363 terminated by !#, just like SCSH. */
364 skip_scsh_block_comment (port
);
365 /* EOF is not an error here */
366 c
= scm_flush_ws (port
, (char *)NULL
);
367 goto tryagain_no_flush_ws
;
370 j
= scm_read_token (c
, tok_buf
, port
, 0);
371 p
= scm_istr2bve (SCM_CHARS (*tok_buf
) + 1, (long) (j
- 1));
378 j
= scm_read_token (c
, tok_buf
, port
, 1);
379 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
384 j
= scm_read_token (c
, tok_buf
, port
, 0);
386 return SCM_MAKICHR (c
);
387 if (c
>= '0' && c
< '8')
389 p
= scm_istr2int (SCM_CHARS (*tok_buf
), (long) j
, 8);
391 return SCM_MAKICHR (SCM_INUM (p
));
393 for (c
= 0; c
< scm_n_charnames
; c
++)
395 && (scm_casei_streq (scm_charnames
[c
], SCM_CHARS (*tok_buf
))))
396 return SCM_MAKICHR (scm_charnums
[c
]);
397 scm_wta (SCM_UNDEFINED
, "unknown # object: #\\", SCM_CHARS (*tok_buf
));
399 /* #:SYMBOL is a syntax for keywords supported in all contexts. */
401 j
= scm_read_token ('-', tok_buf
, port
, 0);
402 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
403 return scm_make_keyword_from_dash_symbol (SCM_CAR (p
));
408 SCM sharp
= scm_get_hash_procedure (c
);
410 if (SCM_NIMP (sharp
))
412 int line
= SCM_LINUM (port
);
413 int column
= SCM_COL (port
) - 2;
416 got
= scm_apply (sharp
,
418 scm_acons (port
, SCM_EOL
, SCM_EOL
));
419 if (SCM_UNSPECIFIED
== got
)
421 if (SCM_RECORD_POSITIONS_P
)
422 return *copy
= recsexpr (got
, line
, column
,
423 SCM_FILENAME (port
));
429 scm_misc_error (s_read
, "Unknown # object: %S",
430 scm_listify (SCM_MAKICHR (c
), SCM_UNDEFINED
));
435 while ('"' != (c
= scm_getc (port
)))
437 SCM_ASSERT (EOF
!= c
, SCM_UNDEFINED
, "end of file in ", "string");
439 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
440 scm_grow_tok_buf (tok_buf
);
443 switch (c
= scm_getc (port
))
469 SCM_CHARS (*tok_buf
)[j
] = c
;
474 SCM_CHARS (*tok_buf
)[j
] = 0;
477 str
= scm_makfromstr (SCM_CHARS (*tok_buf
), j
, 0);
481 case'0':case '1':case '2':case '3':case '4':
482 case '5':case '6':case '7':case '8':case '9':
487 j
= scm_read_token (c
, tok_buf
, port
, 0);
488 p
= scm_istring2number (SCM_CHARS (*tok_buf
), (long) j
, 10L);
493 if ((j
== 2) && (scm_getc (port
) == '('))
495 scm_ungetc ('(', port
);
496 c
= SCM_CHARS (*tok_buf
)[1];
499 scm_wta (SCM_UNDEFINED
, "unknown # object", SCM_CHARS (*tok_buf
));
504 if (SCM_KEYWORD_STYLE
== scm_keyword_prefix
)
506 j
= scm_read_token ('-', tok_buf
, port
, 0);
507 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
508 return scm_make_keyword_from_dash_symbol (SCM_CAR (p
));
512 j
= scm_read_token (c
, tok_buf
, port
, 0);
516 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
522 _Pragma ("noopt"); /* # pragma _CRI noopt */
526 scm_read_token (ic
, tok_buf
, port
, weird
)
532 register scm_sizet j
;
536 c
= (SCM_CASE_INSENSITIVE_P
? scm_downcase(ic
) : ic
);
537 p
= SCM_CHARS (*tok_buf
);
544 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
545 p
= scm_grow_tok_buf (tok_buf
);
552 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
553 p
= scm_grow_tok_buf (tok_buf
);
561 case SCM_WHITE_SPACES
:
562 case SCM_LINE_INCREMENTORS
:
566 scm_ungetc (c
, port
);
594 scm_ungetc (c
, port
);
602 c
= (SCM_CASE_INSENSITIVE_P
? scm_downcase(c
) : c
);
612 _Pragma ("opt"); /* # pragma _CRI opt */
616 scm_lreadparen (tok_buf
, port
, name
, copy
)
627 c
= scm_flush_ws (port
, name
);
630 scm_ungetc (c
, port
);
631 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
633 ans
= scm_lreadr (tok_buf
, port
, copy
);
635 if (')' != (c
= scm_flush_ws (port
, name
)))
636 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
639 ans
= tl
= scm_cons (tmp
, SCM_EOL
);
640 while (')' != (c
= scm_flush_ws (port
, name
)))
642 scm_ungetc (c
, port
);
643 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
645 SCM_SETCDR (tl
, scm_lreadr (tok_buf
, port
, copy
));
648 SCM_SETCDR (tl
, scm_cons (tmp
, SCM_EOL
));
656 scm_lreadrecparen (tok_buf
, port
, name
, copy
)
664 register SCM tl
, tl2
= SCM_EOL
;
665 SCM ans
, ans2
= SCM_EOL
;
666 /* Need to capture line and column numbers here. */
667 int line
= SCM_LINUM (port
);
668 int column
= SCM_COL (port
) - 1;
670 c
= scm_flush_ws (port
, name
);
673 scm_ungetc (c
, port
);
674 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
676 ans
= scm_lreadr (tok_buf
, port
, copy
);
677 if (')' != (c
= scm_flush_ws (port
, name
)))
678 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
681 /* Build the head of the list structure. */
682 ans
= tl
= scm_cons (tmp
, SCM_EOL
);
683 if (SCM_COPY_SOURCE_P
)
684 ans2
= tl2
= scm_cons (SCM_NIMP (tmp
) && SCM_CONSP (tmp
)
688 while (')' != (c
= scm_flush_ws (port
, name
)))
690 scm_ungetc (c
, port
);
691 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
693 SCM_SETCDR (tl
, tmp
= scm_lreadr (tok_buf
, port
, copy
));
694 if (SCM_COPY_SOURCE_P
)
695 SCM_SETCDR (tl2
, scm_cons (SCM_NIMP (tmp
) && SCM_CONSP (tmp
)
699 if (')' != (c
= scm_flush_ws (port
, name
)))
700 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
703 tl
= SCM_SETCDR (tl
, scm_cons (tmp
, SCM_EOL
));
704 if (SCM_COPY_SOURCE_P
)
705 tl2
= SCM_SETCDR (tl2
, scm_cons (SCM_NIMP (tmp
) && SCM_CONSP (tmp
)
711 scm_whash_insert (scm_source_whash
,
713 scm_make_srcprops (line
,
726 /* Manipulate the read-hash-procedures alist. This could be written in
727 Scheme, but maybe it will also be used by C code during initialisation. */
728 SCM_PROC (s_read_hash_extend
, "read-hash-extend", 2, 0, 0, scm_read_hash_extend
);
730 scm_read_hash_extend (chr
, proc
)
737 SCM_ASSERT (SCM_ICHRP(chr
), chr
, SCM_ARG1
, s_read_hash_extend
);
738 SCM_ASSERT (SCM_FALSEP (proc
) || SCM_NIMP(proc
), proc
, SCM_ARG2
,
741 /* Check if chr is already in the alist. */
742 this = *scm_read_hash_procedures
;
746 if (SCM_NULLP (this))
748 /* not found, so add it to the beginning. */
749 if (SCM_NFALSEP (proc
))
751 *scm_read_hash_procedures
=
752 scm_cons (scm_cons (chr
, proc
), *scm_read_hash_procedures
);
756 if (chr
== SCM_CAAR (this))
758 /* already in the alist. */
759 if (SCM_FALSEP (proc
))
762 if (prev
== SCM_BOOL_F
)
764 *scm_read_hash_procedures
=
765 SCM_CDR (*scm_read_hash_procedures
);
768 scm_set_cdr_x (prev
, SCM_CDR (this));
773 scm_set_cdr_x (SCM_CAR (this), proc
);
778 this = SCM_CDR (this);
781 return SCM_UNSPECIFIED
;
784 /* Recover the read-hash procedure corresponding to char c. */
786 scm_get_hash_procedure (c
)
789 SCM rest
= *scm_read_hash_procedures
;
793 if (SCM_NULLP (rest
))
796 if (SCM_ICHR (SCM_CAAR (rest
)) == c
)
797 return SCM_CDAR (rest
);
799 rest
= SCM_CDR (rest
);
806 scm_read_hash_procedures
=
807 SCM_CDRLOC (scm_sysintern ("read-hash-procedures", SCM_EOL
));
809 scm_init_opts (scm_read_options
, scm_read_opts
, SCM_N_READ_OPTIONS
);