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. */
42 /* Software engineering face-lift by Greg J. Badros, 11-Dec-1999,
43 gjb@cs.washington.edu, http://www.cs.washington.edu/homes/gjb */
67 SCM_SYMBOL (scm_keyword_prefix
, "prefix");
69 scm_option scm_read_opts
[] = {
70 { SCM_OPTION_BOOLEAN
, "copy", 0,
71 "Copy source code expressions." },
72 { SCM_OPTION_BOOLEAN
, "positions", 0,
73 "Record positions of source code expressions." },
74 { SCM_OPTION_BOOLEAN
, "case-insensitive", 0,
75 "Convert symbols to lower case."},
76 { SCM_OPTION_SCM
, "keywords", SCM_UNPACK (SCM_BOOL_F
),
77 "Style of keyword recognition: #f or 'prefix"}
80 SCM_DEFINE (scm_read_options
, "read-options-interface", 0, 1, 0,
83 #define FUNC_NAME s_scm_read_options
85 SCM ans
= scm_options (setting
,
89 if (SCM_COPY_SOURCE_P
)
90 SCM_RECORD_POSITIONS_P
= 1;
95 /* An association list mapping extra hash characters to procedures. */
96 static SCM
*scm_read_hash_procedures
;
98 SCM_DEFINE (scm_read
, "read", 0, 1, 0,
101 #define FUNC_NAME s_scm_read
106 if (SCM_UNBNDP (port
))
108 SCM_VALIDATE_OPINPORT (1,port
);
110 c
= scm_flush_ws (port
, (char *) NULL
);
113 scm_ungetc (c
, port
);
115 tok_buf
= scm_makstr (30L, 0);
116 return scm_lreadr (&tok_buf
, port
, ©
);
123 scm_grow_tok_buf (SCM
*tok_buf
)
125 scm_vector_set_length_x (*tok_buf
, SCM_MAKINUM (2 * SCM_LENGTH (*tok_buf
)));
126 return SCM_CHARS (*tok_buf
);
132 scm_flush_ws (SCM port
, const char *eoferr
)
136 switch (c
= scm_getc (port
))
141 scm_wta (SCM_UNDEFINED
, "end of file in ", eoferr
);
145 switch (c
= scm_getc (port
))
151 case SCM_LINE_INCREMENTORS
:
155 case SCM_LINE_INCREMENTORS
:
156 case SCM_SINGLE_SPACES
:
167 scm_casei_streq (char *s1
, char *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)
188 recsexpr (SCM obj
,int line
,int column
,SCM filename
)
190 if (SCM_IMP (obj
) || SCM_NCONSP(obj
))
194 /* If this sexpr is visible in the read:sharp source, we want to
195 keep that information, so only record non-constant cons cells
196 which haven't previously been read by the reader. */
197 if (SCM_FALSEP (scm_whash_lookup (scm_source_whash
, obj
)))
199 if (SCM_COPY_SOURCE_P
)
201 copy
= scm_cons (recsexpr (SCM_CAR (obj
), line
, column
, filename
),
203 while ((tmp
= SCM_CDR (tmp
)) && SCM_CONSP (tmp
))
205 SCM_SETCDR (copy
, scm_cons (recsexpr (SCM_CAR (tmp
),
210 copy
= SCM_CDR (copy
);
212 SCM_SETCDR (copy
, tmp
);
216 recsexpr (SCM_CAR (obj
), line
, column
, filename
);
217 while ((tmp
= SCM_CDR (tmp
)) && SCM_CONSP (tmp
))
218 recsexpr (SCM_CAR (tmp
), line
, column
, filename
);
219 copy
= SCM_UNDEFINED
;
221 scm_whash_insert (scm_source_whash
,
223 scm_make_srcprops (line
,
234 /* Consume an SCSH-style block comment. Assume that we've already
235 read the initial `#!', and eat characters until we get a
236 newline/exclamation-point/sharp-sign/newline sequence. */
239 skip_scsh_block_comment (SCM port
)
241 /* Is this portable? Dear God, spare me from the non-eight-bit
242 characters. But is it tasteful? */
247 int c
= scm_getc (port
);
250 scm_wta (SCM_UNDEFINED
,
251 "unterminated `#! ... !#' comment", "read");
252 history
= ((history
<< 8) | (c
& 0xff)) & 0xffffffff;
254 /* Were the last four characters read "\n!#\n"? */
255 if (history
== (('\n' << 24) | ('!' << 16) | ('#' << 8) | '\n'))
260 static SCM
scm_get_hash_procedure(int c
);
262 static char s_list
[]="list";
265 scm_lreadr (SCM
*tok_buf
,SCM port
,SCM
*copy
)
272 c
= scm_flush_ws (port
, s_scm_read
);
273 tryagain_no_flush_ws
:
280 return SCM_RECORD_POSITIONS_P
281 ? scm_lreadrecparen (tok_buf
, port
, s_list
, copy
)
282 : scm_lreadparen (tok_buf
, port
, s_list
, copy
);
284 scm_wta (SCM_UNDEFINED
, "unexpected \")\"", "read");
291 p
= scm_sym_quasiquote
;
296 p
= scm_sym_uq_splicing
;
299 scm_ungetc (c
, port
);
304 scm_lreadr (tok_buf
, port
, copy
),
306 if (SCM_RECORD_POSITIONS_P
)
307 scm_whash_insert (scm_source_whash
,
309 scm_make_srcprops (SCM_LINUM (port
),
313 ? (*copy
= scm_cons2 (SCM_CAR (p
),
314 SCM_CAR (SCM_CDR (p
)),
324 p
= scm_lreadparen (tok_buf
, port
, "vector", copy
);
325 return SCM_NULLP (p
) ? scm_nullvect
: scm_vector (p
);
346 scm_ungetc (c
, port
);
351 /* start of a shell script. Parse as a block comment,
352 terminated by !#, just like SCSH. */
353 skip_scsh_block_comment (port
);
354 /* EOF is not an error here */
355 c
= scm_flush_ws (port
, (char *)NULL
);
356 goto tryagain_no_flush_ws
;
360 j
= scm_read_token (c
, tok_buf
, port
, 0);
361 p
= scm_istr2bve (SCM_CHARS (*tok_buf
) + 1, (long) (j
- 1));
369 j
= scm_read_token (c
, tok_buf
, port
, 1);
370 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
375 j
= scm_read_token (c
, tok_buf
, port
, 0);
377 return SCM_MAKE_CHAR (c
);
378 if (c
>= '0' && c
< '8')
380 p
= scm_istr2int (SCM_CHARS (*tok_buf
), (long) j
, 8);
382 return SCM_MAKE_CHAR (SCM_INUM (p
));
384 for (c
= 0; c
< scm_n_charnames
; c
++)
386 && (scm_casei_streq (scm_charnames
[c
], SCM_CHARS (*tok_buf
))))
387 return SCM_MAKE_CHAR (scm_charnums
[c
]);
388 scm_wta (SCM_UNDEFINED
, "unknown # object: #\\", SCM_CHARS (*tok_buf
));
390 /* #:SYMBOL is a syntax for keywords supported in all contexts. */
392 j
= scm_read_token ('-', tok_buf
, port
, 0);
393 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
394 return scm_make_keyword_from_dash_symbol (SCM_CAR (p
));
399 SCM sharp
= scm_get_hash_procedure (c
);
401 if (SCM_NIMP (sharp
))
403 int line
= SCM_LINUM (port
);
404 int column
= SCM_COL (port
) - 2;
407 got
= scm_apply (sharp
,
409 scm_acons (port
, SCM_EOL
, SCM_EOL
));
410 if (SCM_UNSPECIFIED
== got
)
412 if (SCM_RECORD_POSITIONS_P
)
413 return *copy
= recsexpr (got
, line
, column
,
414 SCM_FILENAME (port
));
420 scm_misc_error (s_scm_read
, "Unknown # object: ~S",
421 scm_listify (SCM_MAKE_CHAR (c
), SCM_UNDEFINED
));
426 while ('"' != (c
= scm_getc (port
)))
428 SCM_ASSERT (EOF
!= c
, SCM_UNDEFINED
, "end of file in ", "string");
430 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
431 scm_grow_tok_buf (tok_buf
);
434 switch (c
= scm_getc (port
))
460 SCM_CHARS (*tok_buf
)[j
] = c
;
465 SCM_CHARS (*tok_buf
)[j
] = 0;
468 str
= scm_makfromstr (SCM_CHARS (*tok_buf
), j
, 0);
472 case'0':case '1':case '2':case '3':case '4':
473 case '5':case '6':case '7':case '8':case '9':
478 j
= scm_read_token (c
, tok_buf
, port
, 0);
479 p
= scm_istring2number (SCM_CHARS (*tok_buf
), (long) j
, 10L);
484 if ((j
== 2) && (scm_getc (port
) == '('))
486 scm_ungetc ('(', port
);
487 c
= SCM_CHARS (*tok_buf
)[1];
490 scm_wta (SCM_UNDEFINED
, "unknown # object", SCM_CHARS (*tok_buf
));
495 if (SCM_PACK (SCM_KEYWORD_STYLE
) == scm_keyword_prefix
)
497 j
= scm_read_token ('-', tok_buf
, port
, 0);
498 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
499 return scm_make_keyword_from_dash_symbol (SCM_CAR (p
));
503 j
= scm_read_token (c
, tok_buf
, port
, 0);
507 p
= scm_intern (SCM_CHARS (*tok_buf
), j
);
513 _Pragma ("noopt"); /* # pragma _CRI noopt */
517 scm_read_token (int ic
, SCM
*tok_buf
, SCM port
, int weird
)
519 register scm_sizet j
;
523 c
= (SCM_CASE_INSENSITIVE_P
? scm_downcase(ic
) : ic
);
524 p
= SCM_CHARS (*tok_buf
);
531 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
532 p
= scm_grow_tok_buf (tok_buf
);
539 while (j
+ 2 >= SCM_LENGTH (*tok_buf
))
540 p
= scm_grow_tok_buf (tok_buf
);
548 case SCM_WHITE_SPACES
:
549 case SCM_LINE_INCREMENTORS
:
553 scm_ungetc (c
, port
);
581 scm_ungetc (c
, port
);
589 c
= (SCM_CASE_INSENSITIVE_P
? scm_downcase(c
) : c
);
599 _Pragma ("opt"); /* # pragma _CRI opt */
603 scm_lreadparen (SCM
*tok_buf
, SCM port
, char *name
, SCM
*copy
)
610 c
= scm_flush_ws (port
, name
);
613 scm_ungetc (c
, port
);
614 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
616 ans
= scm_lreadr (tok_buf
, port
, copy
);
618 if (')' != (c
= scm_flush_ws (port
, name
)))
619 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
622 ans
= tl
= scm_cons (tmp
, SCM_EOL
);
623 while (')' != (c
= scm_flush_ws (port
, name
)))
625 scm_ungetc (c
, port
);
626 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
628 SCM_SETCDR (tl
, scm_lreadr (tok_buf
, port
, copy
));
631 SCM_SETCDR (tl
, scm_cons (tmp
, SCM_EOL
));
639 scm_lreadrecparen (SCM
*tok_buf
, SCM port
, char *name
, SCM
*copy
)
643 register SCM tl
, tl2
= SCM_EOL
;
644 SCM ans
, ans2
= SCM_EOL
;
645 /* Need to capture line and column numbers here. */
646 int line
= SCM_LINUM (port
);
647 int column
= SCM_COL (port
) - 1;
649 c
= scm_flush_ws (port
, name
);
652 scm_ungetc (c
, port
);
653 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
655 ans
= scm_lreadr (tok_buf
, port
, copy
);
656 if (')' != (c
= scm_flush_ws (port
, name
)))
657 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
660 /* Build the head of the list structure. */
661 ans
= tl
= scm_cons (tmp
, SCM_EOL
);
662 if (SCM_COPY_SOURCE_P
)
663 ans2
= tl2
= scm_cons (SCM_CONSP (tmp
)
667 while (')' != (c
= scm_flush_ws (port
, name
)))
669 scm_ungetc (c
, port
);
670 if (scm_sym_dot
== (tmp
= scm_lreadr (tok_buf
, port
, copy
)))
672 SCM_SETCDR (tl
, tmp
= scm_lreadr (tok_buf
, port
, copy
));
673 if (SCM_COPY_SOURCE_P
)
674 SCM_SETCDR (tl2
, scm_cons (SCM_CONSP (tmp
)
678 if (')' != (c
= scm_flush_ws (port
, name
)))
679 scm_wta (SCM_UNDEFINED
, "missing close paren", "");
682 tl
= SCM_SETCDR (tl
, scm_cons (tmp
, SCM_EOL
));
683 if (SCM_COPY_SOURCE_P
)
684 tl2
= SCM_SETCDR (tl2
, scm_cons (SCM_CONSP (tmp
)
690 scm_whash_insert (scm_source_whash
,
692 scm_make_srcprops (line
,
705 /* Manipulate the read-hash-procedures alist. This could be written in
706 Scheme, but maybe it will also be used by C code during initialisation. */
707 SCM_DEFINE (scm_read_hash_extend
, "read-hash-extend", 2, 0, 0,
710 #define FUNC_NAME s_scm_read_hash_extend
715 SCM_VALIDATE_CHAR (1,chr
);
716 SCM_ASSERT (SCM_FALSEP (proc
) || SCM_NIMP(proc
), proc
, SCM_ARG2
,
719 /* Check if chr is already in the alist. */
720 this = *scm_read_hash_procedures
;
724 if (SCM_NULLP (this))
726 /* not found, so add it to the beginning. */
727 if (SCM_NFALSEP (proc
))
729 *scm_read_hash_procedures
=
730 scm_cons (scm_cons (chr
, proc
), *scm_read_hash_procedures
);
734 if (chr
== SCM_CAAR (this))
736 /* already in the alist. */
737 if (SCM_FALSEP (proc
))
740 if (prev
== SCM_BOOL_F
)
742 *scm_read_hash_procedures
=
743 SCM_CDR (*scm_read_hash_procedures
);
746 scm_set_cdr_x (prev
, SCM_CDR (this));
751 scm_set_cdr_x (SCM_CAR (this), proc
);
756 this = SCM_CDR (this);
759 return SCM_UNSPECIFIED
;
763 /* Recover the read-hash procedure corresponding to char c. */
765 scm_get_hash_procedure (int c
)
767 SCM rest
= *scm_read_hash_procedures
;
771 if (SCM_NULLP (rest
))
774 if (SCM_CHAR (SCM_CAAR (rest
)) == c
)
775 return SCM_CDAR (rest
);
777 rest
= SCM_CDR (rest
);
784 scm_read_hash_procedures
=
785 SCM_CDRLOC (scm_sysintern ("read-hash-procedures", SCM_EOL
));
787 scm_init_opts (scm_read_options
, scm_read_opts
, SCM_N_READ_OPTIONS
);